Thesis. Reference. Methodological approaches for the protection of cultural heritage in the digital age. FONI, Alessandro

Transcription

1 Thesis Methodological approaches for the protection of cultural heritage in the digital age FONI, Alessandro Abstract Les perspectives offertes par les représentations digitales 3D dans les domaines liés à la protection de l héritage culturel posent des questions d ordre méthodologique concernant le type de média et les technologies qui peuvent être utilisées afin d obtenir un résultat qui soit satisfaisant pour tous ceux qui oeuvrent dans ces secteurs. De ce fait, l objectif principal de ce travail est celui d identifier les éléments que ces différents acteurs considèrent essentiels pour qu un projet de visualisation puisse satisfaire aux exigences et aux défis qui caractérisent leurs secteurs d activité. Notre étude déterminera quelles approches, quelles stratégies de création de contenu et quelles technologies chacun de ces groupes perçoit comme étant les plus adaptées afin de répondre à leur besoins respectifs. Cette analyse posera également les fondements conceptuels nécessaires à la définition d un ensemble d éléments convergents et de traits communs qui caractérisent les préférences de ces acteurs : ces convergences pourront être exploitées afin de proposer des représentations [...] Reference FONI, Alessandro. Methodological approaches for the protection of cultural heritage in the digital age. Thèse de doctorat : Univ. Genève, 2013, no. SES 826 URN : urn:nbn:ch:unige DOI : /archive-ouverte/unige:32383 Available at: Disclaimer: layout of this document may differ from the published version.

2 Methodological approaches for the protection of cultural heritage in the digital age THÈSE présentée à la Faculté des sciences économiques et sociales de l Université de Genève par Alessandro Foni sous la direction de prof. Nadia Magnenat-Thalmann pour l obtention du grade de Docteur ès sciences économiques et sociales mention systèmes d information Membres du jury de thèse: Mme. Nadia MAGNENAT-THALMANN, Professeur M. Dimitri KONSTANTAS, Professeur, président du jury M. Jean-Henry MORIN, Professeur M. George PAPAGIANNAKIS, Professeur, Université de Crète Thèse no 826 Genève, 2 décembre 2013

3 ii Methodological approaches for the protection of cultural heritage in the digital age La Faculté des sciences économiques et sociales, sur préavis du jury, a autorisé l impression de la présente thèse, sans entendre, par là, n émettre aucune opinion sur les propositions qui s y trouvent énoncées et qui n engagent que la responsabilité de leur auteur. Genève, le 2 décembre 2013 Le doyen Bernard MORARD Impression d'après le manuscrit de l'auteur

4 Table of Contents iii Table of Contents Table of Contents... iii Résumé... vii Abstract... ix Acknowledgements Chapter 1. Introduction Background: actors involved in cultural heritage Traditional Approaches Advent of New Media and Digital Tools Modern Approaches to Cultural Heritage Protection Problem statement Motivation and pertinence Objectives Organization Chapter 2. Conceptualization The concept of Cultural Property Initial definition of Cultural Heritage item Further classifications of heritage properties Extent of the application range of the Cultural Heritage definition Size of the cultural heritage item Nature of the cultural heritage item Tangibility of the cultural heritage item Cultural significance of the heritage item Categorization of cultural heritage items Dimensions of the protection of Cultural Heritage Documentation Investigation Conservation Defense Dissemination Digitization Groups involved into cultural heritage protection related activities Relation between the groups and their intervention areas Content holders Field personnel Cultural intermediaries Content consumers Digital intermediaries Summary Chapter 3. Related work: taxonomy and review Methodological axes and dimensions Continuum of virtuality Continuum of interactivity... 70

5 iv Methodological approaches for the protection of cultural heritage in the digital age Continuum of precision Continuum of visual consistency Continuum of automatism and span of development cycle Typological classification models for the existing approaches Overview of the identified approaches First quadrant: low virtuality and low interactivity Second quadrant: low virtuality and high interactivity Third quadrant: high virtuality and low interactivity Forth quadrant: high virtuality and high interactivity Mixed approaches Modeling strategies for visualizations exhibiting high virtuality Polygonal modeling D Scanning Image based modeling VRML modeling Spline surfaces Digital photography Summary Chapter 4. Methods Research question Area of investigation Experimental assessment Definition of the targeted groups of interest Surveys Assumptions Hypotheses Questionnaires Dimensions Indicators Independent variables Dependent variables Exploitation of the data Expected results Interviews Objective Size of the sample Preparation and protocol of the interviews Exploitation of the data Expected results Previous work Summary Chapter 5. Analysis Variable recoding and weighting Data aggregation and reliability analysis Technological affinity Interactivity Virtuality Visual consistency and Precision

6 Table of Contents v Automatism Univariate analysis Gender Age Education Domain of activity Technological affinity Frequency distribution of the aggregate scales Bivariate analysis Gender Age Education Technological affinity Domain of activity Preference profiles Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Scale standardization and final preference values Summary Chapter 6. Results and conclusion Ameliorations Results Applications Contributions Limitations Future Work List of publications Bibliography Appendix 1: ACM JOCCH Paper Appendix 2: Questionnaire Appendix 3: SPSS tables

7 vi Methodological approaches for the protection of cultural heritage in the digital age

8 Résumé vii Résumé L utilisation de représentations graphiques, usuellement sous forme de dessins ou perspectives en deux dimensions, constitue une de premières tentatives pour mieux conserver, diffuser et disséminer la documentation relative aux découvertes archéologiques et de manière générale concernant les sites d héritage culturel [OGLEBY01]. La création manuelle de restitutions détaillées constitue un support précieux qui permet une meilleure compréhension du contexte de l époque à laquelle un site a été bâti. De plus, ces méthodes de représentation offrent la possibilité d expérimenter graphiquement et de transmettre des hypothèses spécifiques [SANDERS99] concernant l apparence originale supposée d un site donné. Cela permet donc à la fois de répondre au besoin de documentation du site et, en même temps, crée un support pour l interprétation des découvertes. La démocratisation des technologies de l information, l arrivée récente sur le marché de cartes vidéo GPU à haute performance et la disponibilité de CPU rapides à bas couts, ainsi que les avancées récentes dans le domaine de l infographie 3D, ont permis l essor de nouvelles méthodes de représentation, d analyse, de conservation et de divulgation de la documentation relative aux sites d héritage culturel. Ces techniques offrent des nouveaux supports pour effectuer des études de restitution afin de tester des hypothèses, ainsi que la possibilité de toucher un public plus large [MILLER94] [FORTE97]. Bien que plusieurs projets se basant sur différentes technologies et approches, et visant la réalisation de représentations virtuelles de sites d héritage culturel, aient été réalisées, il n a pas été possible d identifier une méthodologie qui réponde aux attentes et besoins de plusieurs groupes ciblés. Cela a souvent limité l utilité de ces travaux à un seul groupe spécifique d utilisateurs finaux. Par exemple, si d un côté les reconstructions adressées au public général, telles que les films générés par ordinateur, ne sont que rarement considérées favorablement par les archéologues, de l autre côté les reconstructions qui visent en premier les historiens et les archéologues ne sont souvent pas d accès facile pour le grand public [SIDERIS02]. Cependant, les perspectives offertes par les représentations digitales 3D dans le domaine de l archéologie et de la conservation de l héritage culturel [REILLY92] posent certaines questions d ordre méthodologique concernant la forme optimale, le type de média, l approche appropriée et les technologies à utiliser pour obtenir une restitution qui soit satisfaisante à la fois pour les besoins de documentation du conservateur, pour les besoins d interprétation de l historien ainsi que pour les attentes de haut réalisme visuel du public [ADDISON00]. Par conséquent, un des principaux objectifs de ce travail est celui d identifier les éléments essentiels pour les différents groupes actifs dans le domaine de la protection de l héritage culturel et sur lesquels ceux-ci basent leurs décisions de considérer un type de visualisation comme réussie. Par ailleurs, on essayera également de déterminer quelles approches et technologies chacun de ces acteurs perçoit comme étant les plus adaptées afin de répondre à leur besoins et nécessités.

9 viii Methodological approaches for the protection of cultural heritage in the digital age Ainsi, cette étude essayera d établir les bases conceptuelles nécessaires à l identification de lignes directrices méthodologiques claires et de pratiques adaptés qui puissent correspondre aux spécificités des groupes d utilisateurs finaux. Ces lignes directrices, si appliquées, devraient permettre une amélioration de l efficacité, de l utilité et de l impact de futurs projets reliés à l héritage culturel virtuel, puisqu elles devraient faciliter la mise en place de cycles de prototypage plus courts. En effet, la quantité de travail requise pour produire une représentation virtuelle qui réponde aux attentes et aux besoins d un groupe d utilisateur final donné en serait ainsi optimisée. De plus, la possibilité d employer un ensemble d éléments convergents et de traits communs, dans le but de créer une représentation virtuelle polyvalente prévue pour plusieurs groupes d utilisateurs finaux, sera explorée afin de permettre l identification d une approche méthodologique équilibrée et généralisée qui puisse répondre simultanément aux besoins de tous les acteurs ciblés. La détermination à la fois d une stratégie de visualisation, et d une approche de création de contenu, qui pourraient offrir à ces acteurs un compromis acceptable par rapport à leurs préférences, serait un atout de taille. Notamment, cela éviterait d avoir à créer plusieurs versions d une même application d héritage culturel et permettrait ainsi d économiser en amont le temps qui actuellement est dédié à une procédure de recherche par tâtonnement du média et de la forme plus appropriés pour répondre aux demandes de différents groupes d utilisateurs finaux ciblées par chaque nouveau projet entrepris dans ces domaines.

10 Abstract ix Abstract The use of graphical representations, usually in the form of two dimensional drawings or perspectives, constitutes one of the first attempts to better preserve, extend and divulgate the documentation that pertains to archeological, and more generally, cultural heritage sites [OGLEBY01]. Detailed hand-made restitutions represent a precious support which allows for a better understanding of the time dependant context in which a selected site was built. Furthermore, such visualizations offer the possibility to graphically experiment and easily convey specific hypothesis [SANDERS99] regarding the supposed past appearance of a given site, thus satisfying both the need to offer an accessible way to document a site and to offer a support for the interpretation of any finding. The democratization of information related technologies, the recent advent of low cost, high-performance GPU based video cards, the availability of affordable and fast CPUs, as well as the recent advances in the field of 3D graphics have opened new ways and forms to present, analyze, preserve and divulgate the documentation linked to cultural heritage sites. Such modern technologies have offered new supports for the creation of restitution studies for the purpose of hypothesis testing, as well as they have provided the possibility to reach a broader audience [MILLER94] [FORTE97]. Even though several projects that rely on different technologies and approaches, aiming at the creation of virtual representations of heritage sites, have been developed in the past years, no clear methodology has been identified in order to achieve an optimal balance between the expectations and the needs of the all the groups that could have been targeted by those reconstructions, thus limiting the benefit of such work only to one specific group of end users. For instance, the reconstructions that are targeting the general public through computer generated movies rarely encounter the appraisal of archeologists, while on the other hand, restitutions that target primarily archeologists and historians appear often less accessible for the general public [SIDERIS02]. However, the perspectives offered by the use of 3D digital representations in the field of archeological and heritage preservation [REILLY92] raise some methodological issues as well as some fundamental questions regarding the optimal form, the type of media, the correct approach and the technologies that are to be deployed in order to achieve a restitution that could satisfy both the preservationist s need for documentation, the historian s need for interpretation and the public s expectations for high visual realism [ADDISON00]. As such, one of the main purposes of this work is to identify the primary elements that each of the groups that are active in the field of cultural heritage protection regards as being the most important, and upon which it bases its decision to consider a proposed visualization as a successful one or not. Moreover, we equally seek to determine which approaches and technologies each of those actors perceives as being the most appropriate to satisfy their needs and requirements. As a result, our study will attempt to establish the necessary conceptual bases to allow the identification of clear and practical methodological guidelines that match the specificities of the considered end user groups. Such guidelines, if implemented, should then permit an improvement of

11 x Methodological approaches for the protection of cultural heritage in the digital age the effectiveness, usefulness, and impact of future projects related to virtual cultural heritage, as they should allow for shorter prototyping cycles, since the amount of work required to produce a virtual representation matching both the expectations and the needs of a given user group will be optimized. Additionally, the possibility to employ a set of converging elements and common traits to enable the successful constitution of a multi-purpose virtual representation intended for multiple end-user groups will be explored to allow the identification of a balanced and generalized methodological approach that could simultaneously meet the needs of each targeted actor. The determination of a visualization strategy and a content creation approach that could offer such actors an acceptable trade-off regarding their preferences should prove to be a valuable asset to eliminate the necessity to prepare, through a trial and error process, several different virtual heritage applications, on different medias and under different forms, in order to meet the requirement of different end-user groups.

12 Acknowledgements 11 Acknowledgements First of all, I would like to take this opportunity to express my gratitude to Professor Nadia Magnenat-Thalmann for the guidance, the support and the encouragement she continuously provided, as well as for having offered me the privilege to join a gifted and multidisciplinary group such as MIRALab. Her unique perspective as well as her novel ideas supported and inspired me all the while. I equally wish to extend my gratefulness to all the members of the jury, to Professor Dimitri Konstantas and Dr. George Papagiannakis as well as to Dr. Jean-Henry Morin, Dr. Laurent Moccozet and Amal Tawfik: their expert advice, their help, as well as their sharp comments and feedback have been an invaluable source of incentive and motivation. Moreover, I gratefully acknowledge the funding received from the INCO-MED FP5 European Project ERATO (2006) as well as the funding received from the Swiss Virtual Campus projects CATCH and Understanding 3D (2008): they supported preliminary research on some of the topics discussed in this manuscript as well as the creation of most of the 3D models that have been used to produce the renderings included in the questionnaires that have been employed during the data gathering process. Furthermore, I desire to express my thanks to Professor Jean-Paul Descoeudres (University of Geneva), Professor Pierre Sánchez (University of Geneva) as well as to Marc-Andre Haldimann (head of the Geneva museum of archaeology) who preciously contributed to this work by kindly accepting to share their experience and their opinion concerning the themes discussed in the frame of this work. For the same reasons, special thanks are equally due to Cristina Dalla Valle, Marco Fontanini, Dr. Michel Pawlak, Agostino Giaffreda, Massimo Soldini, Nedjma Cadi-Yazli, as well as to Cpt. Roberto Piantoni and the personnel dedicated to the protection of cultural heritage serving in the Civil Protection of the Region of Lugano. Amid both the former and the current members of MIRALab and the CUI, many have helped and supported me through the years on countless occasions and in ways far too diverse to be fully described: suffice to say that their talents, trust, dedication and friendship have greatly contributed to render the time I spent among them an interesting and exciting period of my life filled with exhilaration and enthusiasm. More particularly, I am very grateful to Marlène Arevalo-Poizat, Nabil Sidi-Yacoub and Jean-Claude Moussaly, which have been my mentors before becoming my colleagues and friends. Their guidance, passion and constant encouragement have significantly and positively influenced my academic choices: without them, my latent interest for computer graphics would probably have remained just a hobby. Additionally, I would like thank Camillo Tanzi, Angela Macciocchi and Anna Ferrara, whose confidence in my ability to survive University never wavered, as well as Antonio Latella and Luciano Valsangiacomo, who taught me that studying our past is not just a boring exercise exclusively involving dusty books or neglected ruins, but that it is instead a necessary element to understand our present and a defining aspect of our cultural identity.

13 12 Methodological approaches for the protection of cultural heritage in the digital age Finally, I would like to express my deepest gratitude towards my family and close friends, who never hesitated to push me and kick me, whenever necessary, to better help me put things into perspective throughout the long, difficult, and sometimes sleepless, process that has been the preparation of this work. A special thanks to Margherita and Valentina Foni, my sisters, as well as to Carlo and Tiziana Foni, my parents, who taught me how to appreciate art in all its forms, and who educated me to value, preserve, respect and treasure our cultural heritage: their unconditional patience, reassurances, understanding, love and support have been essential during all these years.

14 Introduction 13 Chapter 1. Introduction From the dawn of time, and during the course of centuries, humankind has always shown the need to build organized groups to survive the hostile environment and to optimize their collective efforts through a rational subdivision of the tasks and duties among the members of their collectivities. This behavior originates from the need of these collectivities to effectively gather the necessary resources to sustain their communities and from the desire to grow stronger from their respective differences. At first these groups where small and disperse and lived primarily hunting and collecting fruits and vegetables naturally found in the environment, later the advent of agriculture and the domestication of animals allowed these communities to grow larger and settle to form the first villages. Later, these collectivities expanded themselves, encountered other neighboring communities and started to exchange their resources extending their network of interrelations. Slowly the necessity to create some kind of superstructure capable to regulate the interrelation between the growing number of individuals and between groups issued from a common region began to rise and different forms of social and political structures started to appear. Eventually some of these communities gave birth to whole civilizations, empires, cities and republics that rose and fell through the course of history, due to natural causes or due to the shock they may have accused clashing between themselves, through cultural assimilation or military conquest. Another peculiarity that mankind has demonstrated to be inclined to, is its necessity to express its creativity, to record the past events in order to better understand the present challenges and to adapt themselves to their environment, or to modify it in order to make it more suitable for their needs. For these reasons, during their lasting, these civilizations evolved, gave birth to exceptional individuals, developed philosophical, legal and religious systems, made fascinating discoveries and invented amazing techniques, built remarkable structures and produced fine works of art. Rarely these civilizations disappeared without leaving any trace of their presence or passage; nor were destroyed completely when they were subjugated by other civilizations. More often, their infrastructure would rather have been integrated, preserved or used to similar or other purposes by the new order in place and, in some cases, the population would have been assimilated in the new social structure through an intercultural bi-directional socialization mechanism. In other terms, on the long run, the adopted and the adopting civilizations could respectively influence each other to the point at which they would share or acquire some of the values, social behaviors or beliefs of the other one. This process would create a more stable society that could benefit from the differences between the two amalgamated cultures and that could reduce possible raising cultural and social tensions. The world as it is today is often the reflection of the marks that all these successive civilizations have left in the course of their development on the trails of history. It is not possible to understand the problems, cultural differences and challenges that we are facing today without the knowledge of the past and of the roots from which our present has directly developed. Our respective cultures, belief systems, philosophical approaches, political systems, technological and scientific knowledge, language and more generally our forma mentis are, in fact,

15 14 Methodological approaches for the protection of cultural heritage in the digital age the legacy of the past civilization that preceded us and upon which our societies emerged and developed. Since we, as individuals, are best defined by our cultural background, more than by our social one, we cannot ignore the lessons and the heritage that was entrusted to us by our ancestors. This legacy was preserved throughout the centuries, generation after generation, and we are bound to transmit it to our descendants as a concrete testimony of the development of humankind and its diverse societies, because without the knowledge of the grounds that made possible the growth of our culture, without our culture, we lose what makes us what we are. In this respect, our cultural heritage, specific to each community, is a precious legacy that encapsulates the origins of the diverse spectrum of possible societies and ideas that humankind has experienced, built or produced, and that even today has a strong influence on our daily life, even at a subconscious level. Our cultural heritage can encompass very diverse items such as our religion, mythology, philosophy, architecture, poetry and literature, artifacts, dialects and languages, paintings or statues. It is evident however, that while all human production can be generally considered as our cultural heritage, due to practical limitations it cannot equally be protected in its entirety. Therefore, since it is not possible to preserve everything that is issued from the human production, the question of the relative importance of the separate items that constitute our collective heritage could be posed. It is clear that some items could be categorized in respect of the degree of importance that they represent for the communities at their different levels or in respect of their rarity. For example, a cultural item that is perceived as important at the level of a certain village, such as an old farmhouse, could be considered less important, if a choice should be made, compared to a cultural item that is perceived as important for a whole country, such as an old roman vestige. More generally, we can therefore define our cultural heritage as any human production that bears witness to a certain stage in the development of a certain civilization and that should be protected for the general interest and preserved for future generations. Our heritage does however not limit itself exclusively to concrete or abstract productions of the human mind. If we are to draw a parallel between the evolution humankind, its civilizations, cultures, societies and the slow evolution and development of our environment, we clearly see that not only some natural places are invested with a cultural significance due the connotations that they may have in respect to certain human cultures, activities or beliefs, but that they possess an inherent importance on their own. Through the course of centuries different civilizations periodically clashed and history has been often characterized by the alternation of periods of war and peace. These struggles, along with natural disasters, have constantly threatened, to a certain extent, the preservation of human cultural heritage due to their inherent destructive power. In the past, natural cataclysms, such as, for example, volcano eruptions, were in this respect more dangerous than the fires of war, and respectively human activities had a limited impact on the natural environment. The advent of the industrialization and the modern times inverted this relation. The scientific advances, applied to warfare and to the civil society, made progressively possible to decrease or prevent to a certain extent the damages that could come from natural causes, but on the other hand, they equally

16 Introduction 15 allowed humankind to create weapons that could match the destruction brought by the most severe natural cataclysm. Moreover, the impact that human activities had on the environment grew more important with the massive urbanization and the atmospheric pollution caused by the growing use of polluting energy sources and chemical compounds employed by the industry. In a relatively short time, humankind managed to reshape and modify, with new tools and techniques, the landscape to better adapt it to its needs. Only recently, we have measured and considered the extent of the negative impacts that such radical changes in our societies had on the development of the natural environment, ranging from climatic changes, which threaten the survival of certain biotopes and generally harm the biodiversity, to the destruction or alteration of unique natural places. Since our diverse natural environment and its uniqueness made possible the appearance of human civilizations and their survival, we should consider it as an essential part of our heritage that we are equally bound to preserve from the dangers our societies may pose, trying to maintain its delicate balance. For these reasons, in respect to the general definition that we used to characterize human cultural heritage, natural heritage could therefore also be defined as any natural phenomenon or manifestation that bears witness to a certain stage in the development of the natural world and that should be protected for the general interest and preserved for future generations. Our collective heritage, either issued from human cultural production or issued from the eminently creative and astonishing work of nature, should then be protected, conserved and preserved to the best of our abilities against the dangers to which it could be exposed: natural disasters, pollution, destruction caused by armed conflicts, vandalisms, pillage, economic speculation, ignorance. In order to preserve the knowledge of the past and promote the awareness of the delicate balance of our environment, different methodologies have been experimented in the past. This approaches range from making written or pictorial reproductions, to establishing specific documentations and archives, to the democratization of the access to education and the instauration of museums and research institutions, to the creation of divulgative essays and informative campaigns targeting the public. The recent development of new digital information and communication technologies and their rapid diffusion among all the different strata of our societies, due to a fast process of democratization to their access, has triggered a profound societal change that affected and modified in a substantial way most of the aspects of our daily life. The emergence of these new technologies, accompanied by an increasing interconnectivity of these information systems on a global scale, has redefined and modified our work habits, our social interactions, our educational system and even our entertaining activities with the appearance of new media. The advent of the digital age and the revolution that modern countries have experienced with the introduction of these new technologies, offers new tools, possibilities and chances for the protection of cultural heritage while providing the main actors that are involved in this fields with new supports to document, analyze and disseminate pertinent and reliable material concerning specific cultural items. Moreover, the introduction of modern 3D visualization technologies, popularized through the increasing diffusion of

17 16 Methodological approaches for the protection of cultural heritage in the digital age video games, the recent advent of low cost, high performance GPU based video cards and CPUs, in conjunction with the recent advances in the field of interactive 3D graphic and scientific visualization, have opened exciting new opportunities of integration and synergy between these technologies and the classical approaches traditionally applied to cultural heritage protection. The scope of this thesis is to analyze the new possibilities and perspectives that are offered to the actors involved in the field of cultural heritage protection by the introduction of these new information and communication technologies. Even though a general presentation of all possible visualization approaches and methodologies, both traditional and modern, will be introduced, our primary focus will be centered on the technologies that are issued from the field of 3D visualization and simulation when applied to cultural heritage items. Our attention will be mainly directed towards the different actors whose work is in direct relation with cultural heritage, towards the specific needs their work involves and towards the expectations they display in respect to the introduction of 3D visualization technologies. One of our central interests is to clarify which opinion they have regarding these technologies, which usefulness they see in their introduction in their specific field of activity and under which conditions and to which extent they are willing to consider them as a valid new tool. An equally important aspect we will explore is whether these new methodologies can represent a positive addition and a real contribution to the usual approaches these actors normally implement in the frame of their activities, and how these new tools can be integrated with the traditional ones they are used to rely upon. This chapter will present an introduction to the themes and elements that will be discussed, analyzed and formalized more in depth in the following chapters. An overview of the actors and approaches that belong to the domain of cultural heritage protection will be presented along a brief summary of the general themes and objectives we seek to investigate in the development of this thesis. 1.1 Background: actors involved in cultural heritage The general definition of cultural heritage, as previously introduced, regroups a wide range of items, ranging from small artifacts and large natural or built sites, to more abstract notions, like specific languages and oral tradition. Likewise, the range of activities, studies, institutions and research fields that are linked, directly or indirectly, at different levels, to the protection, promotion, conservation, interpretation and investigation of this heritage is very broad and regroups diverse actors. Concerning the protection and conservation of cultural heritage, we can mention the role of international organizations, whose mission is to identify world heritage items, to promote and define guidelines and standards for the practical protection and the safeguard of cultural heritage and to instigate the conclusion of treaties, intended for the member states of the international community, to institutionalize the necessity for the protection of cultural heritage at a global scale. Then we can cite the essential involvement of national governments and regional and local authorities, which is indispensable to implement effective policies aimed at the safeguard of our heritage as well as to guarantee the availability of concrete means to achieve this goal of public interest. At last, we

18 Introduction 17 should equally consider the public structures and agencies that could be directly involved with the safeguard of cultural heritage, such as the ones specialized in the cataloguing of heritage sites and items, or the ones that are to intervene when a real danger is menacing those items. Depending on the nature of the threat, we can cite organizations such as, for example, the police, the fire departments or the civil protection, or even private bodies like artisans, architects and restoration specialists. Regarding the interpretation and investigation of cultural heritage, two broad categories, and their different specializations and related fields, are primarily concerned: historians and archeologists. Each of these groups strives to better apprehend the past, to better understand past civilizations, historical periods, their changes and revolutions, either political, social, ideological or technical, and more generally to comprehend the mechanisms, forces, interrelations and mistakes that characterized the development of humankind. Access and analysis of original sources and documentation, written, pictographic or physical, and availability of consistent data is a prerequisite for their work to be carried out in suitable conditions, and their mission relies heavily on the capacity of the previously mentioned groups to provide an adequate protection of our collective heritage. Since these two groups are the wardens of the knowledge we have of our past, and of our capacity to interpret and learn from the lessons we may find in the path that humankind has taken through its evolution, they are equally the most competent to identify and recognize whether a specific item is characterized by any important cultural significance and whether it belongs to our cultural heritage. Thus, the work of such specialists plays a fundamental and essential role in the determination and identification of those items, and their collaboration, supervision, expertise, dialogue and interaction with the other actors involved in the field of cultural heritage ensures the definition of valid, effective and coherent lines of action. In relation to the promotion of cultural heritage we can point out the mission of divulgation, knowledge dissemination and education carried out by entities such as museums, the educational system at all its levels and the public authorities through actions such as awareness campaigns organized either by the press, the media, the action of individuals or by institutions with the publishing of books, essays and papers related to the topic of cultural heritage. More generally, these are the actors that produce, diffuse and divulgate information pertinent to the field of heritage protection. In this respect, first the educational institutions, then the media, due to their coverage and the possibility to reach a large spectrum of the population, are in the best position to raise the awareness of the public towards our cultural heritage. For these reasons, those groups are the ones that could benefit the most from the new synergies between the new information technologies and the traditional methodologies associated with the presentation and divulgation of material related to cultural heritage. Concerning this last point, it has to be stressed that the public, generally supposed to receive the results of the actions and initiatives initiated by the previously mentioned groups, is not a passive element but should be regarded as a full actor, whose needs and expectations should be considered to ensure that the messages convoyed by those initiatives are properly targeted, suitably transmitted and correctly perceived by it.

19 18 Methodological approaches for the protection of cultural heritage in the digital age Finally, if we consider the possible interactions and collaborations between the new digital information and communication technologies and the previously cited groups of interest, we should note the central position that technicians and information and visualization specialists occupy in the frame of such interrelations and synergies. Since these technologies generally require a certain amount of technical expertise to be deployed efficiently, it is clear that the main group responsible for the development of adapted tools to be integrated in the traditional pipelines is represented by the experts and researchers connected with these new fields. One specific area towards which we will particularly direct our attention is the field of computer aided visualization techniques applied to activities and applications relevant to cultural heritage protection, documentation and dissemination. In view of the fact that the groups involved in the area of research and development of new digital tools generally present prototyping cycles of variable length, in order to deliver usable and stable applications, the existence of strong ties, communication channels and feedback mechanisms between those and the previously identified actors must be guaranteed. The most exciting visual application or the most astonishing digital tool intended for cultural heritage, despite being developed with the best possible intentions or employing the most advanced technologies, loses all its meaningfulness and usefulness if the needs of the targeted groups are not properly considered during its development stages. A multidirectional and reciprocal communication between the information specialists and all the concerned actors is therefore indispensable for these new instruments to actually represent a valid option to be integrated alongside the existing traditional methodological tools available to those groups. Likewise, the existence of such consultation mechanisms, and the implication of all the targeted actors and intended users in the conception and development of these new tools, would facilitate a positive reaction, an optimized impact and a better response to the diffusion and adoption of such tools by those groups, triggered by their increased effectiveness and usefulness. 1.2 Traditional Approaches All the groups that are involved into cultural heritage related fields have developed specific tools and methodological pipelines that are suited to apprehend the particular dimensions of a cultural item toward which their mission and work is directed. Some of them may be more interested in the extent of the historical substance impregnating the item, others in the successive possible uses to which it was destined through the centuries: their objectives could therefore be different, ranging from analyzing the object to extract useful information to better understand the past, to preserve it for future generations against possible threats that could compromise it. Although their finalities might be different, the tools employed by those groups to apprehend the dimensions of the concerned item are often similar, even though their use might differ from case to case. The traditional approaches, with which the concerned actors are commonly used to work, and that were adapted to their specific needs, are generally characterized by the presence of strong components of manual, physical or mechanical work, and by the presence of paper based, not reusable, supports.

20 Introduction 19 An additional particularity of such tools is their relative accessibility and their limited flexibility: given the necessary prior intellectual knowledge, the amount of time required to deploy them on the field is generally short and does not necessitate any particular preparation. On the other hand, these tools can subsequently generate a considerable amount of work in the further manipulations and modifications that the collected data could demand. Moreover, these tools profoundly rely on the creation of a direct relation between the observed item and the observer, hence disallowing the possibility to post process or improve the quality of the gathered data without further direct contact with the concerned item. As some examples of such tools in the sphere of cultural heritage analysis, study, documentation and conservation we can cite: the direct on site observation, intervention and manipulation of the concerned items, the constitution of extensive written descriptions, the depiction of sketches and the preparation of restitution drawings, the establishment of architectural, archeological or topographical plans and sections, the creation of visual recordings and the organization of photographical surveys. In order to illustrate to a better extent the abovementioned characterization, we can take as examples directly connected to the field of visualization the case of photography as a tool for documentation, and the use of architectural plans as a tool for intervention and protection. In the specific case of the photographic documentation of a cultural heritage item, given the basic prior knowledge of the manipulation of a camera, no particular preparation is necessary to start the gathering of data: it is sufficient to setup correctly its controls, considering the lighting conditions, arrange, if necessary, some form of artificial illumination, and take the picture. Subsequent to the data gathering process on site, the film is developed and the pictures are printed: such procedure requires a certain amount of time and expertise in order to produce optimal usable results. However, the results of this process could present specific problems, such as, for instance, overexposure, underexposure or the presence of excessive blur, which would require the shooting of new images of the concerned subjects. In some cases this necessity could lead to organizational or logistical complications, since a further session should be arranged on site and the concerned items inspected for a second time. Once the final material meets the fixed requirements, the negatives are used to produce the necessary number of copies for further use. One advantage that printed pictographic documentation presents is that is generally easily accessible by non-specialists and efficiently conveys its information; its disadvantage is that, due to its fragility, it has to be properly conserved, since its support tends to degrade easily. Moreover, in order to produce copies of such material the exploitation of the physical negatives is necessary: their proper conservation is therefore essential to ensure the reusability of the originally collected data. In the case of handmade architectural descriptive plans or drawings, similar remarks can be made. While it is relatively easy to deploy such measures, given the proper talent and the understanding of the necessary notations, it is more delicate to preserve the finalized work, to further modify or rectify it, and to quickly prepare and disseminate the necessary copies to the actors that require them to fulfill their mission. Furthermore, in the specific case of architectural

21 20 Methodological approaches for the protection of cultural heritage in the digital age plans and sections, it is important to note that, due to their inherent technical nature, they could pose some clarity problems in respect to their interpretation and exploitation by non-specialists. This kind of representation, in fact, usually does not convey its information intuitively, and since it could be used as a base, or reference, to prepare evacuation and intervention plans, intended for field personnel responsible for the protection of heritage items in the occurrence of natural catastrophes that would require expedite action, such as floods or fires, the exploration of alternative methods to ensure a better transmission of such information could be constructive. Beside the evident benefits that such tools and methodologies offer, it is clear that they do present some inherent limitations and practical restrictions in their implementation which could be optimized. In this respect, the introduction of new approaches issued from the recent advances made in the development of digital media, communication and visualization technologies could reduce the extent, impact and repercussions of such constraints, while allowing a lessening of the amount of manual and repetitive work that is required to ensure a better reusability and ease of manipulation of the produced results. 1.3 Advent of New Media and Digital Tools The second half of the twentieth century witnessed the rapid development of information related technologies and their fast, generalized diffusion in the occidental civil societies at all levels. The appealing possibilities offered by such technologies, first in terms of data management and numerical computing, later in terms of optimization and automatization of repetitive or time consuming tasks, triggered their progressive migration from the sphere of military applications, to the field of scientific research, to their adoption in the area of commercial applications in order to allow for a better productivity and efficiency. The progresses made in electronic miniaturization, energy consumption and interface ergonomics, and the consequent reductions of manufacturing costs, subsequently facilitated the access to such technologies and assisted their diffusion. The introduction of the first low-cost personal computers produced an acceleration of this phenomenon, marking their transformation into ordinary and necessary tools present in all the aspects of our everyday life, and equally signed the emergence of a first digital revolution. A distinctive trait that characterizes such technologies is the inherent diffusion of digital supports supplanting the traditional ones, and the progressive adoption of the no-paper paradigm. Slowly, the conventional physical paper based format formerly favored to support our written production changed to assume an immaterial digital form. This change offered undeniable benefits, such as for instance the possibility to freely modify a given text without the restrictions characterizing the physical supports. On the other hand, the problem of the preservation of such digital supports and their content, to ensure a long-term persistence of our production, has to be considered. Currently available portable and fixed digital supports present remarkable storage capacities, allowing the condensation of large amount of data, that otherwise could have filled entire libraries, in a relatively small space; however, these supports suffer a certain structural fragility that can compromise their content: such occurrence could

22 Introduction 21 have catastrophic effects due to the high concentration of information they can hold. To prevent such eventuality, the creation of redundant data and the instauration of data replication and duplication mechanisms are necessary. Likewise, the growing acceleration in the development in this field, both hardware and software oriented, caused the emergence of a rapid obsolescence phenomenon striking digital supports and data formats. In this respect, although a digital support can be properly preserved form physical degradation, it is still necessary to preserve as well the means to access and read its contents. At short term, this could be achieved by ensuring the retro-compatibility of older formats in new software, or by transferring the content to other media when the obsolescence concerns the physical support itself and the devices able to read it. At longer term, other solutions must be considered, like for instance hardware emulation or even the complete physical preservation of the original hardware and software used to produce such contents. In view of this concerns, it is interesting to note that although the advent of the digital era has offered like never before the possibility to easily create, manipulate and archive all sort of information and intellectual production, paradoxically has never been so difficult to preserve at long term such digital production from degradation caused by the passing of time. The rapid increase in the computing performances of central processing units and the augmentation of the storage capabilities of digital supports, progressively allowed for the handling of even more complex data. Such evolution marked an extension of the applicative possibilities of informatics, facilitating the introduction of digital tools in a varied array of fields. Thus, due to the increased capacities featured by computerized systems, the digital manipulation of material constituted not only by textual information became achievable. The opportunity to digitally create, archive and modify music, images, videos, and more generally multimedia items, became accessible, encouraged by the diffusion of low-cost dedicated hardware intended to accelerate such particular tasks, and specifically adapted to handle such data, such as, for example, graphical and sound processing units respectively specialized in the treatment of video and audio operations. The diffusion of these new tools, joined with the increasing development of their computational power, equally introduced the possibility to manipulate such large amount of data more interactively while providing an increasing number of operations with the support of real time feedback to the user. These capabilities are particularly interesting when applied to the fields of computer graphics, scientific visualization and more generally to the fields that are aiming at the reproduction and transposition of specific attributes of the physical world and its characteristics into computerized models and synthetic representations for testing and simulation purposes. Such potential coupled with the emergence of three-dimensional visualization technologies and with the advances made in image synthesis and rendering, provided the means to reach a digital representation of reality closer to its original counterpart. Moreover, the synergy between these new tools and the multimedia properties of the newly available supports and digitally exploitable data, has allowed the surfacing of new approaches and opportunities applicable to content creation, manipulation and interaction. At the same time, the possibility to link, layer and encapsulate different types of information to such

23 22 Methodological approaches for the protection of cultural heritage in the digital age content encouraged the development of new forms of knowledge transmission and interpretation, and eventually gave birth to new means of semantic communication and expression. The evolution of digital information technologies and the diffusion of digital supports were accompanied by the progressive development of communication related technologies and their infrastructure. The last decades of the twentieth century have witnessed the outcome of a second digital revolution represented by the extension of the interconnection capabilities between computerized systems at a global scale alongside the development of mobile applications. The rapid development of the Internet, its adoption as preferred support to provide fast access and dissemination of information and services and the low cost availability of broadband channels facilitated the introduction and diffusion of such technologies in the civil society and to the general public; this developments fundamentally modified our social habits concerning the way we relate ourselves to the intercommunication process. The improvements in both spatial extension and transmission speed of networked environments have provided new communication and data manipulation tools that have modified our working environment. The diffusion of instant messaging software, s, shared data repositories remotely accessible and the possibility to establish collaborative environments have extended our capability to quickly and easily reach different people and retrieve pertinent information on a global scale, independently from their physical location. Moreover, the increased interconnectivity of digital networked systems, have equally allowed the sharing and exchanging of decentralized heterogeneous resources between multiple users to be employed for dissimilar purposes. Finally, the convergence between the multimedia capabilities of computerized systems and telecommunication and network related technologies, have both provided new methods to convey, disseminate and access digitally stored reusable resources, and have created new opportunities to reach and interact with larger audiences, hence ultimately contributing to the establishment a of a new medium for content manipulation and transmission. 1.4 Modern Approaches to Cultural Heritage Protection Given the growing possibilities that are offered by the new technologies of information and communication, and the development of new multimedia applications for the digital representation of information, the fields related to cultural heritage preservation, protection and interpretation can clearly benefit from the adaptation and adoption of such new digital tools. Their introduction could then contribute to the establishment of new approaches and methodological pipelines that are particularly adapted to the cultural heritage field. Such new approaches, which are based on the extensive use of digital tools, are generally characterized by the presence of strong components of automatization in the acquisition, treatment and processing of the data relative to the investigated item and by the presence of reusable digital supports to store the gathered information. Additionally, these tools feature a relatively limited accessibility and an extended flexibility: the deployment of such tools on the field requires a good

24 Introduction 23 understanding of their technical functioning, and necessitates the setup of specific conditions to ensure the acquisition and creation of exploitable information. On the other hand, given the appropriate technical competencies, the amount of work necessary to further manipulate or modify such material in its digital form is greatly reduced. Moreover, these tools do not require the creation of a permanent direct relation between the observed item and the observer, therefore allowing the possibility to interactively rectify, modify, enhance, adapt and manipulate the gathered information without further contact with the concerned physical item. Finally, these tools give the opportunity to organize and present heterogeneously gathered data on a unified medium, while allowing a user to simultaneously consult, interact and access a varied combination of different content resources, such as, for instance, textual, pictographic or musical information. As a few examples of the possible application of such digital tools to the sphere of cultural heritage, we can cite: the creation of digital catalogues and databases for the consultation and retrieval of the digitally archived material, the digitization of paper based sources through optical scanning for analysis, conservation studies or long term preservation, the digitization of physical artifacts through laser scanning techniques, the constitution of bi-dimensional or threedimensional CAD models for simulation, hypothesis testing, restitution studies or dissemination purposes, the creation of digital visual recordings and the organization of digital photographical surveys. To better understand the specificities and possibilities offered by the application of digital tools, we can consider, similarly to the development we employed concerning the illustration of the traditional methods, the specific cases represented by digital photography and computer aided architectural drawings. In the specific case of a photographic documentation, the advantages offered by the introduction of digital technologies are manifest. After the on site gathering of the digital pictures, the acquired material is directly exploitable and accessible by different computer applications for varied purposes. It can be stored and linked to a database together with other pertinent information relative to its subject for archiving, or it can be easily manipulated and modified for further use. Even though a correct setup of the shooting conditions should not be disregarded, the possibility to digitally correct and rectify the collected images, through operations such as, for instance, perspective correction, color balancing, exposure compensation or tone mapping, offers a considerable asset. The opportunity to post process such data thus lessens the necessity to reacquire most of the compromised material and therefore limits the need to organize further sessions on site. Moreover, since digital cameras generally feature the capacity to store a much greater amount of images than traditional ones, and offer the possibility to visualize such data immediately after the acquisition, the necessity to carefully control the shooting conditions is significantly reduced. By employing a traditional camera such precautions are clearly required, since the results of the shooting can only be observed after the printing process of the negatives. On the contrary, the use of a digital camera allows for the instant verification of the results: in the eventuality that a specific picture presents excessive flaws, such image can easily be deleted and replaced by a new one. Furthermore, since the

25 24 Methodological approaches for the protection of cultural heritage in the digital age acquired images are stored on a digital support, such data can be straightforwardly distributed and accessed by multiple users: digital copies of the original image files can be made effortlessly if specific users need to modify them for their particular needs. Additionally, several printed copies on physical paper based supports can be produced as well, either of the original data or of its modified versions, without having to go through time-consuming processes. In the case of the preparation of computer aided architectural descriptive plans or drawings, similar remarks can be made. The introduction of computer aided tools provides the possibility to effectively archive, retrieve and modify digitally produced drawn material, and it offers the necessary assistance to expedite the interactive data creation and refinement processes. Lines and surfaces that might be incorrectly placed can be quickly erased and corrected without altering the working support, as it is the case when employing a pen and paper approach to execute the same activity. Moreover, the inherent information present in the digitally produced plans can be extended by the superposition and inclusion of additional data, such as for instance material attributes. In addition, such plans can be transposed, with appropriate tools and with the necessary expertise, into three-dimensional models of the concerned items. Such models primarily present the advantage to offer the possibility to rapidly produce arbitrary sections of the studied object and to quickly create perspective views of its volume and appearance by using real world parameters to simulate a virtual camera. Inversely, the preparation of additional sections and isometric representations manually created with the help of traditional tools would necessitate a considerable amount of time and effort. Additionally, the digitally portrayed item can be interactively visualized under different angles and could be employed to create high quality nearly photo realistic renderings of specific static or animated points of view, producing as results either an interactive application, a set of images or an animated movie. Secondly, three-dimensional rendered visualizations feature a more pronounced accessibility and a better clarity concerning their interpretation and utilization by non-specialists than bi-dimensional plans and section on a paper based support. The exploitation of the newly available multimedia capabilities of modern computers, coupled with the digital visualization possibilities offered by computer graphics, therefore provide, in this respect, more adapted ways to efficiently and intuitively convey pertinent and accessible information regarding the concerned items. Such advantages can be particularly significant when expedite decisions or actions are required from field personnel, which has to access such information to coordinate their activity in order to protect cultural heritage related items from the sudden occurrence of some sort of threat to their integrity. Most of the previously cited methods are clearly intended to produce a computerized representation of real items and their characteristics, especially those that include a visualization component. Some of them are simply a direct transposition of traditional methods with the addition of digital features, as it is the case with architectural drawing: the nature of the task is the same as the one performed with traditional tools, but it is achieved through the use of specialized software while replacing the traditional supports with a computerized interface.

26 Introduction 25 In this respect, it interesting to note that such examples can be classified depending on the varying degree of relation they feature toward the studied item. Certain tools, such as optical scanning, produce as a direct result a digital counterpart of a real object through a digitization process: the original resource still exists and is the essential base for the creation of the digital representation. Other tools, inversely, produce results that are fully digitally created: no intermediary step and little to none physical relation is necessary between the real object and its digital counterpart. This difference clearly appears if we consider the example of photography. One approach consists in employing a normal camera, developing the pictures shot and later scanning the printed supports to transpose them into a digital format; another approach consists in using a digital camera in order to have the pictures directly available in a digital format without requiring the digitization process. This distinction clarifies the double mission that the introduction of digital technologies has to fulfill. First, it has to support the documentation, study, dissemination, protection and conservation of cultural heritage with the creation of new tools and by providing traditional ones with increased capabilities. Secondly, it needs to assist the transition that has to be applied to the already constituted resources, stored or produced on traditional supports, to transpose them to new digital formats and supports. As a concluding remark, it is important to note the central role that information specialists have in the implementation and deployment of such new technologies. As they are in the best position to understand the real possibilities and the inherent limits that such new tools present, their responsibility as intermediaries between the needs expressed by the end users and the development of specific applications that could answer those necessities is indispensable. In order to guarantee an adequacy between those requirements and the proposed solutions, the prior knowledge and understanding of such constrains is crucial: any developed application could otherwise entirely lose its usefulness. Moreover, due to their increased exposure to technological novelties information specialists and researchers find themselves in a privileged position to consider the possible applications that such advances could offer whether transposed to other fields, such as cultural heritage. Since the impulsion to adapt, extend and implement new digital tools issued from the scientific research could as well come from the sphere of informatics as a mean to verify, test or validate newly developed approaches or technologies, the risk to create specific ad hoc tools exploitable only in the context of a given case study exists: in such case, these tools would then be characterized by a limited usefulness and by limited generalization possibilities to other cases. Therefore, in order to establish a coordinated effort and a coherent approach to the application of such new technologies to these fields, the role of supervision and consultation provided by the specific experts, such as, for instance, historians or archeologists, is essential to ensure a proper degree of usability of the created content as well as a suitable degree of adaptation of any developed application to the end user requirements.

27 26 Methodological approaches for the protection of cultural heritage in the digital age 1.5 Problem statement The use of graphical representations, usually in the form of two dimensional drawings or handmade perspectives, along with carefully noted measurements of the considered elements, constitutes one of the first attempts to better preserve, extend and divulgate the documentation relative to archeological, historical and, more generally, cultural heritage items [OGLEBY01]. Moreover, the creation of detailed handmade restitutions represents a precious support that allows for a better understanding of the context in which those items were built or manufactured. These representations equally offer the possibility to graphically experiment, easily convey and formulate possible hypothesis [SANDERS99] regarding the original supposed appearance of a heritage item at the time it came to existence, thus satisfying the need to provide an accessible way to document the item and offering, at the same time, a support for the interpretation of the findings. The democratization and advances in information and visualization related technologies have opened new ways and forms to present, analyze, preserve and divulgate the documentation relative to cultural heritage items, as well as they have both offered new supports to produce restitution studies for the purpose of hypothesis testing, and have provided the possibility to reach a broader audience [MILLER94] [FORTE97]. Nevertheless the perspectives offered by the particular use of 3D digital representations in the field of archeological and heritage preservation [REILLY92], raises some methodological issues as well as some questions regarding the optimal form, the type of media, the correct approach and the specific technologies that can be deployed in order to achieve a result that satisfies both the preservationist s need for documentation, the historian s need for interpretation and the public s expectations for a high visual realism [ADDISON00]. Even though several projects that rely on different technologies and approaches, aiming at the creation of virtual representations of heritage sites, have been developed in the past years, no clear methodology has been identified in order to achieve an optimal balance between the expectations and the needs of the all the groups that could have been targeted by those reconstructions, thus limiting the benefit of such work only to one specific group of end users. For instance, the reconstructions that are targeting the general public through computer generated movies rarely encounter the appraisal of archeologists, while, on the other hand, restitutions that target primarily archeologists and historians appear often less accessible for the general public [SIDERIS02]. 1.6 Motivation and pertinence Given the fact that the domain of cultural heritage can intersect with the respective fields of diverse groups and actors, meeting their interest for dissimilar purposes, it is evident that some kind of common ground is needed to allow a proper communication and a mutual understanding between all the parts involved in order to ensure the establishment of consistent approaches and constructive collaborations. Since the nature of cultural heritage is profoundly interdisciplinary, the definition of a set of adapted and effective methodologies is therefore primarily relevant for projects targeting such domain. The constitution of such methodologies is even more pertinent and valuable when the actors that

28 Introduction 27 are joining their efforts for common projects are issued from very different backgrounds that are characterized by a specialized and dissimilar expertise, as it is the case, for example, for projects seeking to combine new information and visualization related technologies with traditional cultural heritage related fields. The existence of this inevitable gap resides in the fact that all these disparate groups and actors present dissimilar priorities and observe the general issue from their own specific point of view, which is influenced by the different nature of their respective roles, missions and areas of expertise. Moreover, since each of their particular domains features specific needs, as well as it requires proper conceptual models and methodological pipelines that are suited for their goals, it is expectable that each of those actors would have some distinct opinions regarding the possible uses and the inherent limitations about the introduction and application of information and visualization related technologies to their specific field of professional activity. More generally, these groups should present some kind of expectation towards the introduction of these new emerging technologies, and should have some sort of a priori opinion about the possible outcome of their application. The surfacing of such inclinations could originate from either an overestimation, or an underestimation, of the possibilities offered by such technologies, or from a skeptical or overenthusiastic attitude about the real usefulness, potential and efficiency that such new tools could offer compared to the traditionally available ones. This behavior is notably present when the considered technologies are issued from the field of 3D computer visualization. In view of the fact that these technologies are often associated with the sphere of entertaining and video gaming, their introduction in the area of cultural heritage is often regarded by cultural heritage specialists more like an optional gimmick, or at best as an aesthetic addition, rather than a real tool that could offer some tangible advantages. A clear identification, evaluation and assessment of the priorities, needs and expectations of such groups is therefore an essential step in order to allow the creation of practical applications that are specifically suited to answer their requirements, that consider their goals and that could therefore have a positive impact and contribution when deployed beside their traditional tools. Thus, if properly targeted and specifically conceived to broaden the available collection of conventional tools, the implementation of new computerized applications could serve as a valuable extension to the traditional methodological approaches that are employed in cultural heritage related fields. These issues are even more critical, and consequently deserve the utmost attention, especially when the considered applications are intended to benefit more than one group at once for dissimilar purposes. This particular circumstance, which is central to our work, would normally require, in most cases, the development of an amount of variations of the original application equal to the number of end user targeted by it. The necessity to modify the amount and nature of the available information, its method of access, its presentation and support, would then occur due to the need to match the particular requirements and interests that characterize each of those groups, therefore generating the expense of a considerable amount of effort and time in the attempt to adapt either the form or the content of a particular application.

29 28 Methodological approaches for the protection of cultural heritage in the digital age This process of adaptation generally follows a trial and error looping pattern which is triggered by the frequent consultation of external advisors that belong to the groups for which the developed application is intended. The existence of a common language, a reciprocal adequacy in the respective expectations that each group could formulate towards the others, a mutual understanding of the respective fields, and more generally an awareness of the constrains, limits and possibilities that are offered by each of them, is essential to limit the number of iterations inherent to this process. In this respect, the identification of a group of minimum shared common core elements that each of the groups involved considers essential and inalienable for the developed application to be suited for their needs, along with a clear determination of each actor specific needs and requirements, would be an invaluable asset. The establishment of practical guidelines would then allow the optimization of the amount of that is work required for such interdisciplinary projects, reduce the amount of iterations necessary to find a common adapted approach and increase the usefulness of the developed applications. Hence, general-purpose core applications could then be prototyped more easily, and later, if necessary, only the addition of supplementary specific features, needed to meet some particular requirement of certain targeted groups, would be implemented through the preparation of specifically oriented variations of the main application. The main benefit would therefore be represented by the possibility to reuse the core application while limiting the amount of further work to the extension of its features: this would minimize the necessity to rebuild from the ground up new elements, and the occurrences where a modification of old elements is required in order to shape them into a different form that is more suited to be used for the purposes of a new field of application. However, if such set of minimum shared requirements does not happen to exist, due the presence of excessively divergent views and specificities that characterize the main actors, the sought guidelines would nonetheless constitute a meaningful tool to properly plan and optimize the development of computerized application intended for a less extended set of end users, or even for only one. The developed application would then integrate the maximum set of common converging elements that the targeted actors deem the most important; then it could be further customized to meet the specific needs of each of the particular groups involved in the project in those cases where they present strongly diverging requirements. Whereas only one actor would be involved in the development of such projects, beside the information and visualization specialists responsible for the technical implementation, a precise and reliable assessment of its needs and expectations would anyhow represent a great advantage for the development of a functional application that is suited for its purposes and that is specifically tailored for its sphere of activity. 1.7 Objectives This thesis aims at identifying the primary elements that each of the groups that are active in the field of cultural heritage protection regards as being the most important, and upon which it bases its decision to consider a proposed visualization as a successful one or not. Moreover, we equally seek to determine

30 Introduction 29 which approaches and technologies, as well as for which reasons and under which circumstances, each of those actors perceives as being the most appropriate to satisfy their needs and requirements. By means of such an analysis we aim to produce clear practical methodological guidelines that account for the specificities of the targeted end user groups, hence allowing a maximization of the effectiveness and the usefulness of future virtual heritage related projects. Through the identification of the core elements that the different selected end user groups consider primarily responsible for a successful virtual representation of an heritage item, i.e. one that conveys all the necessary and pertinent information that matches both their identified needs and expectations, we aim at establishing useful and practical guidelines that will improve the effectiveness, the usefulness and the impact of future applications in the field of virtual cultural heritage. The developed guidelines will serve as a base to both accelerate the development phase for the virtual applications and to optimize the amount of work that is required to produce a virtual representation that matches both the expectations and the needs of the targeted end user groups. Furthermore, we aim at identifying a set of converging elements to be employed for a multipurpose virtual representation that simultaneously targets multiple user groups, and which maximizes the presence of the elements that each group considers critical while avoiding the elements that are considered incompatible with their needs. Consequently, the identification of such a balanced generalized methodological approach, would eliminate the necessity to prepare, through a trial and error process, multiple versions, on different media and under different forms, of a given virtual reconstructions in order to meet the specific expectations of each actor involved or targeted by a given virtual heritage application. 1.8 Organization In this chapter, we have offered an overview, along with some background information, of the main issues, thematic directions and general objectives that are important for the further development of our work. Chapter 2 provides a description and a conceptualization of the specific elements that pertain to our study. A clear definition of the different dimensions relative to cultural heritage, the extent of the application of its definition, the actors involved in its sphere along with their missions and objectives will be presented. Furthermore, the particular dimensions that are relevant to our analysis will equally be emphasized. In chapter 3, the methodological approaches that are applicable to the different visualization techniques in the field of cultural heritage, as well as the related work pertinent to such area, will be categorized and analyzed, alongside the presentation of some significant examples. Advantages and drawbacks of each approach will be considered as well. Chapter 4 covers the methodology applied to our work: research questions, field of application, specific objectives and expected results will be presented. Moreover, the practical implementation of such aspects will be described alongside the modalities and strategies that have been employed to gather the data that will be analyzed, commented and discussed in chapter 5. Finally, a summary of the

31 30 Methodological approaches for the protection of cultural heritage in the digital age results, the limits and future perspectives related to our work, accompanied by some concluding remarks, will be offered in chapter 6.

32 Conceptualization 31 Chapter 2. Conceptualization In this chapter, all the relevant elements, terms and concepts that will be later used for our analysis are considered and characterized through an identification of their pertinent dimensions. The concept of cultural heritage item will be defined and the extent and range of its application presented. Furthermore, the actors involved into cultural heritage related activities and their interrelations will be identified alongside their field and level of intervention. In order to better apprehend such concepts, we will first give a brief overview of the legal dispositions and definitions that are applicable to cultural heritage and its protection. Subsequently, we will present the institutions and organizations that are concerned by those dispositions and that are directly involved in the effort of preservation and protection of such heritage. After the delineation of such general background information, we will then analyze and categorize, through the identification of their inherent properties accompanied by concrete examples, the different items to which it is possible to apply the definition of cultural heritage. Finally, we will highlight the main activities and actions that pertain to the protection of cultural heritage items along with the groups of actors that are concerned, associated or active in those areas. The modification of the relations between those groups, triggered by the introduction of new tools issued from the field of information, communication and visualization technologies, alongside the new role that information specialists are required to play, will be emphasized as well. 2.1 The concept of Cultural Property Cultural heritage items generally exhibit a considerable connection with the evolution and the traditions associated with the communities that either produced them or inherited them, and they represent a unique and irreplaceable element that defines our cultural identities as they bear witness to the development of our civilizations. The comprehension of the specific cultural significance that is featured by such properties thus provides the chance to better understand our past and offers the opportunity to better apprehend the realities of our present world. Cultural properties, due to their inherent invaluable nature, are usually exposed to a wide range of specific dangers that can threaten their integrity, such as armed conflicts, illegal transfers or deliberate destruction, and due to their uniqueness they should be properly protected against the catastrophic effects of external endangering factors, such as natural disasters. The protection of cultural heritage is therefore an important priority for every country, both regarding the safeguard of the cultural properties present on their soil and concerning the encouragement of practices aimed at promoting a general respect towards the items owned by other countries [OFC98]. The international community and all its actors should therefore deploy a collaborative effort to support the establishment of effective measures intended for the reinforcement of the protection and safeguard of endangered cultural items, either featuring an historical, an artistic or a social significance.

33 32 Methodological approaches for the protection of cultural heritage in the digital age To achieve the instauration of efficient means aimed at the preservation of our cultural heritage, and allow the transposition and concretization of such intentions into an adapted legal frame that regulates and supports such activities, it is first essential to apprehend and identify the exact nature which is featured by a cultural property. Hence, clearly recognizing the distinctive traits that can characterize any cultural item is a necessary step to define the precise scope and extent of the regulations and interventions that are susceptible to affect such items. Likewise, determining the application range of the concept that underlies the definition of cultural heritage property is a fundamental element to ensure the deployment of adapted measures for their safeguard and to allow the creation of a suited context that promotes and assists the execution of the activities aimed at their protection. Proceeding to the identification of such aspects, preparing a classification of the traits that define the items exhibiting a cultural relevance and clarifying the conditions under which a property can be considered as pertaining to our cultural heritage, could however resort to a delicate exercise. It could be, in fact, unwise, due to the extreme variety featured by the diverse typological array exhibited by cultural heritage properties, to rely merely on a single definition to encompass all the possible items which might display a cultural significance. Consequently, applying a too strict, limited or rigid definition of heritage property might imply the exclusion of a large amount of items whose cultural significance and contribution to our cultural identity is valuable and undeniable, and whose pertinence for the deployment of protection measures is unquestionable. Opting for a too general, wide or abstract definition of the features that characterize a cultural item might inversely contribute to extend such definition to objects whose cultural relevance is negligible, irrelevant or inexistent, hence causing a dispersal of the resources that are allocated for the preservation of such properties and therefore weakening the overall impact of the protection measures that are intended for their safeguard. In order to better explain and apprehended such flexible concept and the complexities that intervene in its application for the identification of the aspects that characterize any heritage property, we will therefore initially present, accompanied by some background information of historical nature, the definitions and classifications concerning the cultural items as implemented in the first international multilateral conventions related to the promotion of the protection of such properties. Such initial partial definitions, due to the motivations underlying their elaboration, will be then expanded to highlight the full extent of the application range of the concept of heritage item. With the development of further considerations (section 2.2), proceeding from the analysis of specific elements that appear in additional elaborated conventions, which inherently reflect the surfacing of recent concerns and conceptions regarding the nature of heritage items, we will attempt to give a more comprehensive view of the aspects that cumulatively and complementarily define and complete a general concept of heritage property Initial definition of Cultural Heritage item For a long time, in the occurrence of armed conflicts, the pillage and destruction of heritage properties has been a common strategy intended for the humiliation, destabilization and discouragement of the enemy forces. The destruction of the

34 Conceptualization 33 Mostar Bridge, the demolition of the Buddha s statues in Afghanistan and the spoiling of the museums during the wars in Iraq, represent some recent examples of the dangers that cultural heritage items can incur in times of conflict. Between the XVI and XVIII century, the works and writings of philosophers such as Hugo Grotius, John Locke and Jean Jacques Rousseau discussed and presented the firsts theoretical aspects, approaches and doctrines regarding the practices that should be applied to cultural properties in such eventualities [PBC04]. Later, at the Congress of Vienna in 1815, organized after the Napoleonic wars, during which many art treasures were subtracted to adorn the Musée Napoleon, whose name later became the Louvre Museum, partial restitutions were imposed to the French delegation. Such cases, and the prohibition of the subtraction of heritage properties during armed conflicts, agreed during the Vienna congress, might be considered some of the first instances of multilateral decisions concerning cultural restitution practices in world history [SKRYDSTRUP04]. As a further precursory example, we may consider the set of rules prepared by Francis Lieber, and later approved by President Lincoln, aimed at governing the conduct of soldiers in the field during the American Civil War ( ). The Lieber Code, as it is generally referred, stated that in the absence of military necessity the appropriation of cultural treasures in enemy territory could not be justified. Such document became the legal ancestor to the XXth century conventions regarding such practices, as well as to the acts of restitution subsequent to World War II [SKRYDSTRUP04]. However, the first real international legal instruments specifically aimed at the safeguard of cultural heritage properties were first adopted near the end of the XIX century. The 56 th article of The Hague convention of 1899 and 1907 stated that the protection of certain cultural items, against acts of confiscation, indebt appropriation or spoiling, ought to be guaranteed in the frame of war practices [LTBC01]. Moreover, as a further example of the surfacing of such considerations and concerns toward heritage properties, we can mention the case of the Roerich Pact of 1935, whose conclusion involved ten American States that introduced for the first time the obligation to designate valuable heritage properties with distinctive recognizable markings [PBC04]. After World War II, and the vast human and material losses that were caused by such large conflict, the awareness toward the fact that cultural heritage properties are indeed invaluable and unique, encouraged the international community to create international instruments and specific sets of rules explicitly intended for the reinforcement of the protection of such significant items [OFC98]. Thus, the UNESCO was founded in Paris in The UNESCO subsequently promoted works that ultimately culminated with the ratification of The Hague s convention of 1954 [UNESCO54], and its first and second additional protocols, respectively prepared in 1954 [UNESCO54b] and 1999 [UNESCO99], in order to complete the pre-existing legal tools in the frame of the protection of the cultural heritage in the occurrence of armed conflicts. Such convention was acknowledged and ratified, as of 2013, by 126 Countries, and often constitutes the model, inspiration and primary reference for all the regulations and legal dispositions concerning such themes both at international or national level, as it is the case, for instance, for the UNESCO convention of 1970 [UNESCO70], for the Unidroit convention of 1995 [UNIDROIT95], for the

35 34 Methodological approaches for the protection of cultural heritage in the digital age European directive 93/7/CEE [EEC93], or for some Federal and cantonal text of law adopted in Switzerland [LTBC01]. The main fundamental concept appearing in The Hague s convention of 1954 is represented by the will to promote the respect toward the cultural properties situated within the territories of the countries engaged in an armed conflict, by refraining them to use such properties, and their immediate surroundings, for any purpose which could be likely to expose them to destruction or damage (art 4 paragraph 1). Furthermore, it prohibits, any form of theft, pillage, misappropriation and acts of vandalism directed against any cultural property, and discourages the requisitioning of any movable cultural item situated in the territory of any another country (art 4 paragraph 3). A further aspect that exhibits a central role in such convention is the desire to support the efforts directed toward the safeguard and the preservation of the integrity of heritage properties in times of peace against the foreseeable effects of an armed conflict, by establishing the necessary measures that each country considers appropriate (Art.3), and by promoting the employment of distinguishable markings to identify any valuable properties to which such practices should be applied (art. 6). The main point toward which our interest is primarily directed concerns however the initial formalized definition that appears in such convention, which represents the first attempt to clearly determine and apprehend the nature featured by a cultural heritage property alongside its characterizing traits. The scope and object of the application of The Hague s convention of 1954 is highlighted, in fact, in its fist article, which generally presents cultural properties as being specific items, either movable or immovable, that feature a great importance to the cultural heritage of every people. Such definition is then extended (Art.1 paragraph 1) to specifically encompass monuments of architecture, art or history, whether religious or secular, archaeological sites, groups of buildings or works of art, such as manuscripts, books or collections, that might present any historical, archaeological, scientific or artistic interest. Such definition includes as well (Art.1 paragraph 2), all the buildings whose main and effective purpose is to preserve or exhibit movable cultural properties, such as museums, large libraries, depositories of archives, or refuges intended to shelter, in the event of armed conflicts, such items. It is clear that this initial definition presents certain flaws in respect to the type of items that are susceptible to be concerned by its application range. Its inherent limitations are in part due to the underlying intentions and context motivating such convention, which specifically targets the items that are threatened by the occurrence of armed conflicts, and is in part due to the traditional preconceptions existing at that time regarding the exact nature and relative significance of cultural properties. It is, in fact, evident that the primary distinction operated by the definition appearing in The Hague s convention regarding cultural properties mainly concerns their capacity to be transported or relocated. This distinction is significant when the dangers menacing the integrity of cultural items are either issued from warfare activities, such as bombings, or are connected to illegal transfer practices, such as pillage. It is however likely that such restricted definition, and the narrow array of dimensions which are considered to characterize the concerned cultural items, which proceeds from the identification of the potential dangers that might affect them, coupled with the

36 Conceptualization 35 further capacity for each country to subjectively determine the properties which might be encompassed by such definition, contributes in the reduction of its general application range. It is therefore necessary, in order not to exclude from the general concept of cultural property specific items that should be included into its field of application, to extend such first initial definition to embrace a larger spectrum of properties. All the items that might benefit from a widening of the scope featured by such initial attempt to conceptualize and characterize heritage properties are thus not exclusively comprised in the categories represented by movable or immovable objects, and do not necessarily only pertain to the typology expressively established in the text of the convention. A further effort to expand the conceptual frame determining the varied features and the diverse traits exhibited by heritage items is hence required to allow the proper understanding of their multidimensional nature and their full relevance. Likewise, such effort equally represents a fundamental aspect to permit the comprehension of all the pertinent dimensions that suitably identify such items, and to allow the consequent development and deployment of adapted measures aimed at guaranteeing their safeguard both in times of conflict and peace Further classifications of heritage properties During the first years of the 1960s, Mexico and Peru, alarmed by the fact that the phenomenon of illegal transfer and contraband of cultural items and objects of art was rapidly growing to alarming proportions, urged the UNESCO to assist in the preparation and adoption of an adapted international instrument aimed at promoting specific measures intended for the safeguard of cultural heritage properties in times of peace, and expressively suited to answer such concerns [LTBC01]. Subsequently, the General Assembly adopted in 1964 a resolution prohibiting the importation of properties exhibiting a cultural value whenever they were not featuring a properly certified authorization issued from the country from which such item was originating. Moreover, this resolution encouraged all the member countries to collaborate in order to eradicate the practices concerning the illegal transfer of cultural heritage properties [UNESCO64]. With the support of the International Council of Museums (ICOM) and the International Institute for the unification of the Private Law (Unidroit), UNESCO later prepared a report on the potential measures to be implemented to assist these efforts, and its General Assembly charged a group of experts with the responsibility to elaborate a project of convention to achieve such end. The idea to regulate and to coordinate the efforts deployed by the international community to effectively limit, and possibly eliminate, the phenomena associated with contraband of cultural items was concretized in 1970 during the 16 th UNESCO General Conference in Paris with the adoption of the final text of such convention [UNESCO70], which has been ratified, as of 2013, by 124 member States. The UNESCO convention of 1970 represents the first universal agreement concerning the international transfer of cultural properties and aims both at reinforcing the existing legal measures intended for the protection of our cultural heritage and at promoting the safeguard of such valuable properties through the mutual cooperation and collaboration between member States at international level [LTBC01]. Such convention presents the basic principles and sets the

37 36 Methodological approaches for the protection of cultural heritage in the digital age minimal dispositions, both legal and administrative, that each State ought to implement in order to ensure a proper protection and safeguard of the heritage properties against the effect exerted by illegal transfer activities, while leaving enough liberty to each country to adapt such formulations to their own conceptions, specificities and to the possibilities offered by their national legislation [OFC98]. The dispositions included in the text of the convention do not feature however a self-executing capacity, disallowing their direct application and therefore necessitating their prior transposition into the national legislations in order to become effective. Similarly, they do not specify any instrument or mechanism specifically intended for the restitution and repatriation of exported, subtracted or stolen cultural properties. Nonetheless, it should be noted that the 1970 UNESCO convention can be considered, to a certain extent, as a first chart embodying the basic principles for the protection of cultural heritage items in periods other than the ones characterized by armed conflicts. In this respect, its content and dispositions are aimed at completing, specifying and extending the determination of the application range of the initial definition of heritage property appearing in The Hague s convention of Due to its underlying intention and scope, the 1970 UNESCO convention faced the necessity to attempt a more exhaustive identification of the traits characterizing a cultural property in order to allow an objective determination of all the items that its dispositions were targeting. Since the previous definition presented an excessively limited typology of items that were considered as cultural properties, such definition was largely widened and expanded for the finalities of the convention in order to include several other valuable properties. The determination of the concept of cultural property, as adapted in the frame of such convention, is therefore presented through the development of a definition characterized by two main aspects, which aim at establishing a complete list concerning the properties that ought to be considered as part of our common cultural heritage, along with the necessary conditions regarding such acknowledgement. The first part of the definition implies that in order to consider a property as exhibiting a cultural importance, either on religious or secular grounds, it must be first specifically designated by each State as displaying a particular significance either for archaeology, prehistory, history, literature, art or science (art. 1. preamble). Such definition, however, is largely vague and undetermined, since the specific meaning associated with the featured significance is completely entrusted to the interpretation given by the State attempting to protect the cultural property. The convention does not give any further explanation, clarification or suggestion concerning the correct interpretation that might be applied to better define and objectively apprehend the dimensions characterizing a significant item, and therefore creates a first flaw represented by the inclusion of such subjective element participating in the definition of cultural property. Nevertheless, it must be noted that the desire to associate a cultural property to a State (further specified in art 4), thus establishing a link between a given Country and a property exhibiting a national cultural importance, is a necessary aspect required for the deployment of specific measures included into the convention (art. 5, 9, 12 and 14), which are only applicable if a cultural property

38 Conceptualization 37 effectively belongs to national cultural heritage recognized by a member Sate [OFC98]. The second part of the definition appearing in the UNESCO convention of 1970 states that the identified cultural properties must belong, either exclusively or cumulatively, to at least one of the eleven categories listed in art.1 and hereunder reproduced. Such classification exhibits an extremely wide and varied range of items that are considered for the needs relative to the application of the convention, and hence considerably extends the initial definition of heritage property as presented in the 1954 convention. Among the items constituting such first attempt to constitute a comprehensive classification of all the properties displaying a cultural relevance, we can find: o Rare collections and specimens of fauna, flora, minerals and anatomy, and objects of paleontological interest. o Properties relating to history, including the history of science and technology and military and social history, to the life of national leaders, thinkers, scientists and artist and to events of national importance. o Products of archaeological excavations or discoveries. o Elements of artistic or historical monuments or archaeological sites. o Antiquities more than one hundred years old, such as inscriptions, coins and engraved seals. o Objects of ethnological interest. o Properties of artistic interest, such as: o Pictures, paintings and drawings produced entirely by hand on any support and in any material. o Original works of statuary art and sculpture in any material. o Original engravings, prints and lithographs. o o o o o Original artistic assemblages and montages in any material. Rare manuscripts and incunabula, old books, documents and publications of special interest (historical, artistic, scientific, literary, etc.) singly or in collections. Postage, revenue and similar stamps, singly or in collections. Archives, including sound, photographic and cinematographic archives. Articles of furniture more than one hundred years old and old musical instruments. It is clear that such categorization does imply the determination of an application range that is excessively large and dispersed, thus potentially extending the cultural heritage definition to encompass virtually any item due to the absence of objective discriminatory criteria contributing to the establishment of some sort of prioritization and order of relevance amid the objects susceptible to pertain to a given category. Moreover, it shall equally be mentioned that, due to the intended scope of the convention, the proposed categorization obviously targets mainly rare or antique movable objects, without extending the definition of heritage property to include a more varied array of items featuring a different typology in their exhibited traits or nature. Hence, such classification, and the concepts constituting the base for the definition of its categories, can be considered as a continuation and prolongation of the first basic distinction operated by the 1954 convention

39 38 Methodological approaches for the protection of cultural heritage in the digital age between movable and immovable properties. Nonetheless, despite the partial criticisms that can be formulated towards the relative inadequacy of the wide application range that proceeds from the definition of heritage item as it appears in the 1970 convention [O KEFFE89] [GORDON71], it should be noted that, in order to allow such convention to become an effective global instrument, some compromises were inevitable. Due to its not self-executing nature, in fact, it was necessary to carefully consider the inherent cultural diversities and specificities featured by the States potentially adhering to it, thus reducing the possibility to promote the adoption of too strict or narrow definitions, which could have limited their capacity to adapt such text to their respective national legislations. Moreover, the extreme extent of the application range of the 1970 definition might be explained by the considerable amount of aspects that were included in the convention to ensure the fulfillment of its intent, whose application necessitated the determination of a large typology in order to cover the whole spectrum of items concerned by such measures. A further example of text attempting to define the concept of cultural heritage property through the establishment of a comprehensive classification of the typology exhibited by cultural items is represented by the European directive 93/7/CEE on the return of cultural objects unlawfully removed from the territory of a Member State [EEC93]. Adopting essentially the same approach featured by the first part of the definition proposed by the UNESCO 1970 convention, thus entrusting the identification of the significant items to the discretion of each State, it similarly employs a list of fourteen categories (annex A of the directive) to clarify and determine the application range of its dispositions. Furthermore, it introduces in its annex B a discriminatory objective criterion to limit the extension of its proposed categorization in order not to extend the application range of its definition of movable heritage properties to encompass irrelevant or inappropriate items. The directive states, in fact, that a cultural object, besides belonging to at least one of the fourteen previously identified categories, might be subject to the application of the directive only if its value, expressed in ecus, corresponds to, or exceeds, specific established financial thresholds determined for the category it pertains. Such values might range, for instance, from zero ecus, for archeological objects, dismembered monuments, incunabula and manuscripts, to ecus for pictures. Such initial attempts, alongside the creation of comprehensive classifications and categorizations aimed at the identification of the general traits exhibited by any property that could potentially feature a cultural relevance do however suffer, to different degrees, from the opposite basic flaw exhibited by the restrictive and limited definition introduced by The Hague s convention of Consequently, their application range often presents a too large, undefined, vague or excessively wide extension to clearly and objectively apprehend the multiple pertinent dimensions and different aspects that can participate in the determination of the cultural significance exhibited by any valuable heritage item. Likewise, the elements of subjectivity introduced in the processes of identification of such relevance, coupled with the established conceptions regarding the nature and typology pertaining to heritage properties, which proceeds from the evident and traditionally accepted distinction operated between mobile and immobile objects, ultimately prevents such definitions to

40 Conceptualization 39 fully and exhaustively apprehend the general concept characterizing cultural heritage properties. Finally, we may note that adopting such approaches, in order to either achieve a single definition applicable to all heritage items, or attempt the creation of a detailed classification of all the objects to which such concept can be applied, it can amount to a less than practical, or even impossible, task, due to the obvious drawbacks which are implied in either cases. Hence, a too strict, simplistic or limited definition will inevitably exclude significant items whose importance for our heritage is unquestionable, and, inversely, a too large, extensive or generic classification might improperly extend the application of the concept of heritage item to properties whose relevance is questionable or even inexistent. In order to achieve a suitable formulation of such multidimensional concept, to clarify the range of items that might be subject to its application and to elucidate the multiple factors that participate in the characterization of the nature and typology that a cultural item can display, the conceptual elements constituting the different interdependent aspects relative to its definition should therefore be carefully analyzed. Proceeding from the necessity to separately consider such aspects, and from the need to establish and highlight more complex and flexible distinctions, which are required to better identify an adapted typology exploitable to define and describe the extent of the application range of the cultural heritage definition, we will hence present in the following subchapter a breakdown of the distinctive traits we feel as the more pertinent and functional to characterize an heritage property. The additional dimensions that will be considered, and the associated considerations that will be further developed to extend the initial concept of heritage item, as it emerges from the first definitions previously presented, will utilize as a support some specific UNESCO conventions that were adopted after The content of such conventions reflects recent concerns and new conceptions relative to the aspects that might define a valuable heritage property: their presentation will then be illustrated by some concrete examples. Finally, as a concluding remark, we would like to stress that, in order to reduce the impact of any subjective element participating in the determination of the cultural significance of any potential heritage item, which represents a central factor in the definition of valuable heritage property, the role played by cultural specialists, such as historians, coupled with the expertise they can provide to facilitate the identification of any objective trait exhibited by such properties, will undoubtedly represent an invaluable and essential asset contributing to the deployment of appropriate measures aimed at their protection. 2.2 Extent of the application range of the Cultural Heritage definition In the first part of the twentieth century, the concept of valuable cultural heritage item was usually associated with monasteries, churches or ecclesiastic edifices and religious objects. Such definition was generally considered applicable as well to fortifications, castles or ancient vestiges [PBC04]. It is in this respect that the initial definition of cultural property was formulated in The Hague s

41 40 Methodological approaches for the protection of cultural heritage in the digital age convention of 1954 [UNESCO54]. However, the necessity to extend such limited definition to encompass other pertinent items, which were formerly not considered as objects featuring any notable historical significance, slowly emerged. As consequence, an increased attention and awareness was directed towards the cultural importance of items such as civil architectural constructions, industrial complexes, old farms, bridges and old transportation vehicles due to the fact that they can equally be considered as important witnesses to a specific historical period. It is interesting to note that the extension of the concept of cultural heritage to such items redefines as well the temporal threshold we are used to consider in order to decide of the cultural relevance of a particular item. It is generally accepted that old or antique items possess an inherent historical or archeological substance that should be preserved as cultural testimony, However, recent artistic productions or outstanding modern architectural constructions, such as for example Le Corbusier s edifices or Andy Warhol s creations, could as well be considered to a certain extent as cultural properties, independently from their creation period or their antiquity. Furthermore, the concept of cultural heritage could equally be extended to include items that are not necessarily merely limited to physical artifacts or built structures, extending the range of application of such concept to cover immaterial heritage or natural sites as well. It is clear that the remarkable variety in the nature of the items susceptible to be considered as a valuable cultural property engages and involves a diverse array of dissimilar fields that are concerned by their protection, preservation and study at different levels, depending on the relevance that those items represent for their specific area of interest and activity. In this respect, the multiplicity of the inherent distinctive traits that can characterize a cultural item entitles the involvement of specialists issued from different fields, ranging from archaeology, history, architecture or literature, to biology, ecology, geology or linguistics, depending on the degree of suitability they feature to apprehend the relative importance of a particular item and its specificities. Thus, in order to comprehend to a better extent the vast diversity of the application range of the cultural heritage concept and the complexity of the interrelations and necessary collaborations between these diverse areas of expertise, a general classification of the different attributes and qualities that distinguish a cultural item should therefore be attempted Size of the cultural heritage item A first important dimension that is to be considered in order to characterize a cultural heritage item is its size. Since the concept of cultural property does apply independently from the spatial extension and occupied volume of a given item, this attribute is to be taken into account in order to deploy the appropriate tools and measures to protect it and to preserve it in a suitable manner. In respect to the 1954 Hague s convention, the definition of heritage property is applied without restriction to small and medium sized items and works of arts, like for example books, potteries or paintings, such as ancient Roman artifacts, Leonardo s Gioconda or Donatello s David statue, to medium and big sized buildings and monuments, such as Amiens Cathedral or the Taj Mahal, to large architectural complexes and museums, such as Bellinzona s Castles or

42 Conceptualization 41 Versailles, to extended groups of buildings or whole parts of a city or a site, such as Petra, the archeological site in Olympia, the historic center of Florence, Venice and its lagoon or Mt. S. Michel and its Bay [UNESCO08]. It is clear that the disposition that are to be taken when considering the protection and conservation of such heritage items significantly differ from case to case depending on its dimensions. Similarly, the adapted tools and the professional skills that should be deployed to apprehend the particular aspects of a given item that is to be preserved heavily depend on this factor. Hence, specialists ranging from the field of architecture, urbanism, or history to experts in the areas of archeology, history of art and literature may be required to properly classify and pertinently document a specific cultural property. A second dimension that must be considered in relation with the spatial extension of a heritage item is its mass. Such attribute is eminently important when the integrity of the cultural property is threatened and protective measures must be put into place in order to prevent the degradation or destruction of the concerned heritage item. The initial definition of cultural heritage property, as it appears in the first Hague s convention of 1954, states, in fact, a clear distinction between mobile and immobile heritage items. Such distinction is profoundly essential in the preparation of evacuation plans intended for field personnel in the eventuality that a sudden intervention on site may prove necessary. Based on such difference, it is then possible to decide whether it is appropriate to extract the concerned mobile items, such as paintings for example, to dispatch them to a safer storage area, or to rather prepare beforehand the required material to build on site suitable protective structures to shield the endangered immobile items, such as for instance fixed statues or architectural elements Nature of the cultural heritage item A fundamental aspect that undoubtedly defines a cultural property is its inherent nature. As previously introduced, it is essential to fully consider as important constitutive elements of our collective cultural heritage that are worth preserving not only the intellectual and material productions issued from the human activity, but also the wonders issued by the slow work of nature alongside their delicate balance. Since during the last decades cultural heritage properties and the general natural environment were increasingly threatened not only by the traditional causes of decay, but also by the quickly changing social and economic conditions, along with the industrial development and the rapid increase of environmental pollution phenomena, the establishment of a clear definition of both human cultural heritage and natural cultural heritage is essential alongside the general acknowledgement of their absolute importance for our societies. In this respect, considering that the deterioration and disappearance of any item belonging to those heritages constitutes a harmful impoverishment of the heritage of all the nations of the world, the Paris convention of 1972 [UNESCO72] was defined in order to raise awareness towards such issues. In this convention a redefinition and a generalization of the applicability of the concept of human heritage, as mentioned in The Hague s convention of 1954 is proposed, and a broader designation of heritage property that explicitly includes natural elements as integral part of the world heritage is established. Hence, in

43 42 Methodological approaches for the protection of cultural heritage in the digital age such extended definition of cultural heritage, new elements, such as inscriptions, cave dwellings or combinations of features that are of universal value from the point of view of history, art or science, make their appearance. Moreover, a meaningful inclusion into such definition (art. 1, paragraph 4), is represented by the addition of particular elements, such as groups of buildings which, because of their architecture, their homogeneity or their place in the landscape, are of universal value from the point of view of history, art or science, and of sites, works of man, or the combined works of nature and of man, which are of universal value from the historical, aesthetic, ethnological or anthropological points of view. In this respect, for the first time, terms such as landscape, concepts such as the work of nature and dimensions such as the aesthetical or anthropological one, are unambiguously mentioned and their significance properly taken into account in order to enrich the traditional definition of heritage property. Furthermore, in its second part, which is specifically intended for this purpose, the convention clearly extends the initial concept of cultural heritage stating the fist definition of natural heritage. In this particular definition, the natural cultural heritage is identified as: features consisting of physical and biological formations which are of universal value from the aesthetic or scientific point of view; geological and physiographical formations or delineated areas which constitute the habitat of threatened species of animals and plants valuable from the point of view of science or conservation; natural sites or precisely delineated natural areas of universal value from the point of view of science, conservation or natural beauty (art. 2, paragraphs 1 to 3). To better illustrate the extension of the range of application of this new definition of natural cultural heritage, we can cite as examples: the Wet Tropics of Queensland, Mount Athos, the botanical garden in Padua, The landscape of Val d'orcia near Siena, the Yellowstone National Park or the Jungfrau-Aletsch- Bietschhorn region in the Swiss Alps [UNESCO08]. It is important to note that the broadening of the general definition of heritage item, as the inclusion of such new aspects in its initial concept, both expands the area of action and augments the diversity of the tools that are to be used in order to effectively ensure an adapted protection and documentation of such properties. The involvement of a wider array of specialists, providing the necessary expertise and the availability of new skills and tools, specifically tailored to apprehend the particular dimensions characterizing natural cultural heritage, is therefore necessary. Hence, the deployment of the knowledge pertaining to fields ranging from ethnology, geography or geology, to fields such as ecology or biology, alongside the traditional fields related to human cultural properties, is an essential aspect in the effort to preserve efficiently our collective cultural heritage Tangibility of the cultural heritage item The cultural properties considered up until this point are either issued from the human activity or are issued from the natural environment, as defined in the previous sections, and usually present a self-evident tangible and physical nature. These properties can either be directly observed or measured, and can be apprehended by a direct relation with their physical manifestation. However, it

44 Conceptualization 43 is important to equally consider the existence of other elements that also pertain to our common cultural heritage and that are not directly observable and that do not present a clearly identifiable tangible nature. Thus, the generalized definition of cultural heritage item should be extended to include in the list of important items worthy of protection for the general benefit new elements that can be qualified as intangible. These items represent the abstract creations of the human mind that cannot be transposed into a physical and fixed tangible form, and that are deeply interconnected with the development and creation of the cultural properties pertaining to the physical human cultural heritage. For instance, we can consider traditional tales, legends and religious practices and beliefs as integral parts of the cultural background of specific societies, and ancient and modern languages or dialects, particularly in the case of ethnic minorities, as an important constitutive element of the cultural identity of the communities that are defined by them. Likewise, the surviving traditional performances pertaining to local folklore, as any preserved antique theatre play, for instance, can also be considered as living witnesses of the ancient times that must be conserved and protected for future generations. These surviving immaterial testimonies also represent a precious asset to better understand the reality of those periods in which they were commonplace, and consequently are of undeniable social and ethological value. The importance and necessity to define the immaterial cultural heritage emerged during the Paris convention of 2003 [UNESCO03]. This convention was specifically intended to stress the significance of such specific dimension of our common cultural heritage, and directly defines the intangible cultural heritage as the practices, representations, expressions, knowledge, skills, as well as the instruments, objects, artifacts or cultural spaces that are associated with them, that communities, groups and, in some cases, individuals recognize as part of their cultural heritage. This intangible cultural heritage, transmitted from generation to generation, that is constantly renewed by the communities and the groups in response to their environment through their interaction with nature and in relation with their history provides them with sense of identity and continuity. The intangible cultural heritage, as defined in the convention (art.2 point 2), expressively includes and manifests itself in domains such as oral traditions and expressions, including language as a vehicle of the intangible cultural heritage, performing arts, social practices, music, rituals and festive events, knowledge and practices concerning the nature or the universe, and traditional craftsmanship. In this respect, Roman theatres were constructed as social gathering places hosting public representation of oral performances; churches, cathedrals and mosques were built under the impulsion provided by the faith shared by the communities that erected them as gathering places for the expression of their beliefs and the Egyptian burial sites were structured to reflect essential elements of the Egyptian religion and mythology. The existence of such remarkable tangible built heritage properties is thus a resulting manifestation of immaterial intangible elements that are at their origins. Preserving and understanding the intangible elements and the knowledge that inspired and allowed the creation of such heritage items, is therefore as

45 44 Methodological approaches for the protection of cultural heritage in the digital age important as to conserve and protect the physical vestiges built by past civilizations and the physical artifacts that are considered as cultural properties Cultural significance of the heritage item Regardless of the nature, spatial extension and tangibility of a cultural heritage item, the main characterizing aspect that has to be considered when cataloguing, studying or preparing appropriate protection measures intended for their safeguard, is their relative importance. As previously introduced, several general distinctions between significance categories, with their associated intervention priorities, can be applied to categorize and classify the cultural properties depending on their cultural relevance. These classifications are generally used to define the application range of specifically adapted legal dispositions affecting definite groups of items that share a common importance at international, national, regional or local level. Thus, more rigorous safeguarding measures, and the responsibility associated with their application, are generally affected at the corresponding level of authority, ranging from national to local, to be properly coordinated. Likewise, the establishment of such distinction is essential in order to set pertinent priorities when the simultaneous safeguard of all the items affected by a sudden threat is not possible and a choice is to be made to protect or to abandon certain artifacts. Due to their extended range of application, and for evident practical reasons, the definition of these item categories is generally organized in the form of successive containers that inherit the parent s relevance class, i.e. any object located inside a structure considered as eminently valuable might inherit such status even though it is not featuring the same cultural relevance, therefore limiting the possibility to clearly distinguish between the relative significance of specific items located at the same level in this hierarchical classification. Since it is primordial to assign priorities at each item level, an objective evaluation based on the characterizing dimensions that determine the item s cultural significance has to be prepared. This process can raise some interesting considerations such as how it is possible to determine whether a particular item is to be considered as a cultural property, whether or not it is sufficiently significant to be included, and to which extent, in the cultural heritage general definition, and ultimately if it presents any cultural relevance whatsoever. Therefore, in order to evaluate such aspects, several additional inherent distinctive traits pertaining to the studied item should be taken into account. The first aspect that is essential to evaluate in order to determine the objective significance of a specific item is its historical importance in relation either to the history of the place where it was located or in relation to the historical qualities featured by the object itself. Thus, structures such as the Berlin wall, which is representative of the historical period of the cold war and of the divided Germany, or objects, such as a medieval plate armor, which is representative of the battle practices of a specific historical period, could both be considered as cultural properties. Secondly, in order to distinguish between the respective significance of similar items featuring comparable historical qualities, their relative, or absolute, uniqueness or rarity should be defined as well. Sites such as Stonehenge, or the

46 Conceptualization 45 statues of the Easter Island, for instance, feature a high degree of absolute rarity and uniqueness in respect to other megalithic sites: no similar or comparable structure exist anywhere else. Objects such as the seventeenth century British crown jewels feature as well a relative uniqueness and rarity: despite the fact that other comparable or similar items exist, they however present a significant historical value, and can be considered unique, due to the physical or symbolic relation they had with specific historical figures, institutions or historical events. A further aspect that plays a central role in the definition of the importance of a cultural property, particularly in relation with the relative uniqueness and rarity of given object, is its integrity, the persistence of its historical substance and eventually its capacity to embody the representative nature and the principal characteristics of a class or type of items. Despite the fact that many partial examples of Roman pottery can be found in several museums, a complete roman amphora, for instance, featuring a good preservation state, could be considered as relatively unique, rare and particularly valuable in respect to its capacity to display its original historical substance, its remarkable structural integrity and its ability to represent the original appearance of the particular category of ancient artifacts to which it belongs. Likewise, the expressionist paintings of Vincent van Gogh, as well as the cubist architectural buildings designed by Josef Chochol in Prague, can also be considered valuable elements that belong to our cultural heritage, due to the fact that they constitute a representative, still existing, example of important historical artistic movements. Besides the abovementioned attributes that could help in the identification of the extent of the cultural significance that a heritage item presents, i.e. historical relevance, uniqueness, integrity and representativeness, additional elements should equally be retained to broaden the range of application of the cultural property definition to cover other relevance categories. In this respect, we could include in the list of aspects that are to be considered when evaluating the cultural importance of a particular item supplementary dimensions such as, for example, its potential to educate, illustrate or provide further scientific investigation in relation to the existing cultural heritage, its capacity to exhibit a good design or specific aesthetic characteristics, and its aptitude to display a particular richness, diversity or an unusual integration of structural features. Furthermore, a cultural significance could as well be featured by objects, places or documents that demonstrate or that are being associated with significant scientific or technical innovations and achievements [NTA02]. Thus, historic places such as brothers Wright s first flight trail area or scientific instruments such as William Crookes firsts experimental spinthariscopes, used to count alpha particles emitted by radium, are representative of the origin of particular advances achieved by humankind in the pursuit of scientific knowledge and therefore constitute an important witness for the history of science. The cultural properties that specifically pertain to the natural heritage, even though not being related to human activity, i.e. not featuring any particular religious, symbolic, historical, ethnological or social attribute, may nevertheless present a significant cultural importance per se. The criteria that can be employed to determine the existence and the extent of such cultural significance

47 46 Methodological approaches for the protection of cultural heritage in the digital age are partly an adaptation of those previously employed to characterize the items featuring a human component, and are partly specific to the particular essence of natural items. Among the distinctive traits that determine the cultural relevance of natural formations we can mention, as a first element in parallel with the main attributes defining the human heritage, their uniqueness and rarity on a provincial, national or global scale. Secondly we can consider their integrity, defined as the extent to which the area is unaffected by human development. Finally, we can cite their representativeness and amount of overall diversity, defined as the extent to which the concerned areas represent the biological and physical features of the considered ecosections (sub-regional areas with minor physiographic, microclimatic or oceanographic variations [ECS-BC05]), their richness, and in particular their ability to adequately capture the full range of biologic, geographic and climatic features of the given areas in their original or actual state [SRM-BC96]. Furthermore, two specific additional dimensions are to be taken into account when the safeguard of a natural environment, landscape, area or ecosection pertains to the preservation of the natural cultural heritage: their long-term viability and their overall vulnerability. It is essential in fact, in order to develop or to deploy specifically adapted and efficient measures to ensure the protection of those particular areas, to evaluate both the realistic ability that is offered to intervene, the amount of resources, constrains and conditions that are required to ideally maintain in perpetuity those areas, and the existing risks of permanent loss that their biophysical elements or features could incur if they were not protected. In view of these aspects, natural formations such as the Grand Canyon, due to its exceptional natural configuration, specific biotopes and ecosystems, such as the ones that can be found in the Galapagos island region or at the Great Barrier Reef on the north-east coast of Australia, are to be fully considered as eminently invaluable heritage properties that ought to be safeguarded and protected due to their biodiversity, beauty, integrity and outstanding uniqueness Categorization of cultural heritage items The variety in the distinctive traits that characterize the inherent attributes featured by the items that are susceptible to be considered as heritage items calls for an adapted implementation and choice of the appropriate tools and methods that are to be deployed to preserve and protect them. In this respect, it is unlikely that the tools that could be suitable to apprehend the specific dimensions of a large natural heritage site would be directly applicable to small heritage human artifacts. Hence, each of the item categories previously introduced, as presented in figure 1, will benefit from the expertise of particular groups of specialists, apt to properly determine the extent of their cultural value, as well as from the deployment of the traditional conceptual and technical tools available to each of those groups in their respective fields. For example, a biologist would be more apt to determine the cultural value of a natural reserve, while the evaluation of an archeologist would be more adapted to define the extent of the cultural relevance of the vestiges of an ancient Roman settlement. However, this distinction is not absolute and the combined effort and collaboration of experts issued from more than one field, thus providing a more

48 Conceptualization 47 extended area of expertise, is often necessary to suitably capture the whole range of cultural significance that a specific item presents. Significance Cultural item: historical, cultural or symbolic relevance, uniqueness, integrity and representativeness Inherent Nature Natural heritage Human heritage Tangibility Tangible Intangible Size, spatial extension Big Medium Small Big Medium Small Figure 1: classification of the items susceptible to pertain to the cultural heritage definition The introduction, application and the form that the integration of digital tools can assume in the traditionally established pipelines belonging to the fields of expertise that could be called to intervene in relation to a given heritage item, significantly differs and should be adapted to both the inherent distinctiveness featured by the concerned item and to the specific needs and requirements displayed by the groups that are involved in its protection and preservation. In relation to the abovementioned examples, the digital tools that could be deployed, either transposed or inspired by the traditional existing ones, or issued from new technological approaches, to apprehend and study a large natural site, such as satellite imagery or computerized simulations of the evolution of the existing ecosystem, do require the involvement of different technologies that the ones necessary, for instance, for the digital documentation of a small cultural artifact. However, due to the data sharing possibilities that are offered by new digital technologies it would be feasible, conversely to what it is possible with traditional approaches, to reuse and integrate the data gathered by the exploitation of tools intended for larger areas, such as digital geographic representations obtained by laser measurements, with the digital information pertaining to smaller items, as it could be the case for geo-referenced databases that were to include digital representations of the cultural items and buildings present in such areas, along with other textual information linked to them. While the application and adaptation of digital tools, and new technologies in general, can be considered for the full spectrum of properties to which the definition of heritage item applies, such as for instance the digital recording or cataloguing of intangible heritage items, such as endangered dialects or tales and dances issued from the local folklore, the scope of this thesis is mainly directed towards the group of items that benefit from a direct implementation of

49 48 Methodological approaches for the protection of cultural heritage in the digital age visualization related techniques. More specifically, the aim of this work is to investigate how the groups that could intervene, either directly or indirectly, in the frame of the visualization, reconstruction, presentation, dissemination, protection, conservation and analysis of cultural items issued from the built tangible human heritage, such as, for instance, churches, artifacts, statues, groups of buildings or archeological sites, respond to the introduction of such new techniques. 2.3 Dimensions of the protection of Cultural Heritage Independently from the distinctive traits that a cultural heritage item presents, and from the place it occupies in respect to the classification of its defining features, the actions and missions relative to its protection, that are carried out by the groups that are best suited to apprehend its significance, follow several different axes of intervention. The general concept of protection of cultural property presents a multidimensional nature, which becomes apparent when all the different facets that constitute the generic notion of protection are simultaneously explored and applied through synergic approaches and exchanges of information among the groups that involved into heritage related activities. The general conception of heritage protection can be divided into different interdependent complementary aspects and activities that express their full potential impact when deployed collaboratively. In this respect, it can be noted that the introduction of digital technologies facilitates and expedites the exchange, coordination, sharing and reuse of the information that is required to achieve such end. The first aspects that are to be mentioned are represented by the creation of comprehensive documentations for the concerned heritage items and the investigation and analysis of their characteristic features. Subsequently, as further dimensions that could be cited, we can refer to the deployment of adapted conservation measures to limit their degradation and the preparation of protection procedures that are intended for the safeguard of their integrity against external dangers. Finally, we can also mention the dissemination of the information relative to their significance, which can raise the awareness of the public toward such cultural properties by spreading the knowledge concerning our cultural heritage: these activities would in this respect serve as an educating tool. Each of these axes of intervention contributes in its way, in its specific sphere of action and with its specific tools and approaches, which are related to its particular mission and purposes, to reinforce the global effort of preservation of the cultural heritage as a whole. When combined and deployed jointly to apprehend the vast range of complex aspects that characterize a cultural property, these approaches represent optimal means to give a comprehensive representation of the multidimensional nature that is featured by cultural items. In the following sections, we will briefly present the different abovementioned dimensions and axes that are connected with the general concept of protection of cultural heritage, i.e. documentation, investigation, protection, conservation and dissemination. Additionally, some examples about how these dimensions relate to the protection of the tangible built human cultural heritage and about

50 Conceptualization 49 the pertinent digital tools that are relevant for the achievement of the objectives that are associated with a given axis will be presented as well Documentation The task to constitute a comprehensive and detailed documentation of a given heritage item is the first and indispensable activity upon which the other axes of intervention linked to the concept of protection of cultural property rely. It is, in fact, necessary for such actions to be properly planned and executed to benefit from consistent and certified information about the different aspects and characteristics that a cultural item presents. The main role of the informationgathering mission, in order to fulfill its documenting function, is to extensively catalogue and systematically archive all the pertinent data that can facilitate the identification and determination of the specific features that characterize a particular item and to provide the appropriate means to quickly retrieve and communicate such information. The type of the gathered data, ideally aimed at establishing an all-inclusive description of the full range of distinctive traits that are associated with the multidimensional nature exhibited by cultural items in general, can vary from textual physical descriptions, to pictographic, photographic and video recordings, to structural diagrams. The collected data, however, should not limit itself exclusively to the geometrical attributes of a concerned item, and should include external information such as historical references, evaluations on its observable preservation conditions, cross references to other cultural items and specific documentation that is pertinent to its specific nature, such as geographic surveys in the case of natural cultural items. Concerning tangible human cultural heritage, this activity is embodied by the deployment of traditional tools, such as the shooting of photographic documentations, the preparation of architectural plans and sections, the evaluation of the physical measurable dimensions that are featured by the item, such as, for example, size, weight, constitutive materials, localization, and by the gathering of the historical references and cultural information that could support the evaluation of the item s cultural significance. The digital counterparts of the tools that are employed for the documentation activities, include the deployment of technologies and applications such as computerized scanning techniques, the exploitation of computer aided drawing solutions, in order to define the geometrical properties of a given object, the employment of digital imagery related approaches, to constitute visual documentation of the appearance of the concerned items and the utilization of electronic databases to store, organize and retrieve the digitally gathered information Investigation The accurate, and scientifically rigorous, investigation of cultural items is an essential aspect that is connected to the general protection of our common cultural heritage. This activity occupies, in fact, a central role in the objective determination of the cultural significance of the considered cultural items, and provides the information necessary to properly establish their relative value and

51 50 Methodological approaches for the protection of cultural heritage in the digital age the contribution that they represent in respect to the existing cultural heritage as a whole. Hence, the investigation effort provides the essential factual data that is required to allow a correct evaluation of adapted protection measures that are to be deployed to ensure the long-term preservation and protection of the concerned heritage items. The assessment of the cultural qualities and inherent structural characteristics, as opposed to the formal observable ones that are associated with a cultural property, heavily depends on the qualitative and quantitative data gathered by the documentation process. Proceeding from such collected information, the investigation task aims at determining the extent of the inherent cultural significance of the studied object as well at apprehending its contribution in the broadening of the knowledge linked to the already existing heritage properties; moreover, it assists the efforts that aims at furthering the understanding of our past, our cultural roots, origins, specificities and ultimately of our cultural identity. Furthermore, the analysis of cultural heritage items can support the elaboration of effective and specifically adapted conservation measures that are intended for endangered items through the identification of any jeopardizing effects that external phenomena could exercise on their integrity, such as, for instance, pollution, humidity and climatic changes, which could threaten, directly or indirectly, their preservation and their existence. Amid the traditional tools that can be deployed to assist the investigation activity we can cite as examples: historical, archeological or architectural surveys, aimed at determining the cultural interrelations and historical significance of the concerned items and the creation of restitution studies, either in the form of drawings, scale models or physical reconstructions, intended both to support the interpretation of the existing surviving data and to widen the existing knowledge relative to specific aspects of the studied cultural object, its typology or the historical period to which it belonged. Further elements that can be mentioned in such context are the organization of live experimental archeology experiences and the analysis of a cultural property structural and physical constitutive attributes. Such analysis contributes to the evaluation of its preservation state and to the determination of the causes its possible degradation, as well as to the identification of future potential threatening factors, to the estimation of the risks of permanent loss it may sustain if not properly protected and to the assessment of effective conservation practices aimed at preventing the occurrence of such eventualities. Concerning the tools issued from the implementation of computer related technologies to such ends, we can cite: the utilization of electronic measurement and testing instruments, such as image spectrographs, electronic microscopes and image processing techniques, as well as the creation of computerized models and simulations, in order to investigate, apprehend and understand the time dependant-evolution of specific aspects that are affecting the cultural property in order to consider any preemptively appropriate solution for their preservation. Moreover, among the digital tools that are relevant for the study of cultural properties, we can also mention the implementation of 3D visualization technologies and the manipulation of three-dimensional representations as

52 Conceptualization 51 instruments to rapidly produce digital restitution studies for hypothesis testing and for validation or simulation purposes: these studies constitute a virtual transposition of the experiences that are usually associated with the field of experimental archeology and that can provide the necessary context to support the interpretative and analytical tasks Conservation The efforts that are aimed at preserving the existing cultural heritage are a vital task that ensures the long-term transmission of our cultural legacy to its future inheritors. The conservative interventions, based on the established documentation, proceeding from the information gathered through a proper study of the concerned heritage items and supported by the contribution issued from the investigation of effective methods to protect the cultural property from external endangering phenomena, are aimed at maintaining and safeguarding such items from the degradations caused by the effects of time. To achieve such objective, the actions that are aimed at defending the integrity of cultural properties can range from the implementation of suitable protective measures to prevent any potential degradation to occur, to the deployment of purely conservative dispositions to maintain the actual state of the cultural item, to the application of restructuring interventions aimed at either restituting an appropriate integrity to an endangered item in order to improve its capability to sustain the inevitable aging process or at reverting, if possible, its actual conditions to its original ones without affecting its historical substance. It is to be noted that the abovementioned conservative actions can be initiated by diverse actors at different levels as the result, for instance, of the implementation of political decisions, such as the resolution to officially consider a given site as a protected one and therefore to provide all the necessary means to achieve such goal, or as the result of physical direct interventions that are intended to consolidate the structure of the considered cultural item. As a few examples of interventions that are susceptible to be applied for the preservation of tangible human cultural heritage we can mention: the direct conservative restorations of the historical substance that is exhibited by cultural items such as statues, buildings, paintings and frescoes, as well as the nonconservative interventions, or rebuilding, of items partially or totally damaged beyond repair. An example of such non conservative action is exemplified by the baroque Frauenkirche in Dresden, built between 1726 and 1743, destroyed by the Allied bombing raids at the end of World War II and recently rebuilt on the basis of historical photographs and plans of the original building, employing, whenever possible, the salvaged materials and surviving elements that used to belong to the original structure. The first type of interventions is generally deployed to restitute the original appearance to the damaged, or endangered, items in order to reinforce their resistance to external threatening phenomena or to recover their original magnificence for cultural or dissemination purposes. The non conservative approach is often implemented when a culturally significant item is irremediably lost and when it is considered that its tangible existence in the physical and cultural landscape, even if reconstructed and deprived of most of its historical

53 52 Methodological approaches for the protection of cultural heritage in the digital age substance, is in the interest of the community since it serves a cultural, symbolic or commemorative purpose. As a further activity that can support the preservative efforts that are aimed at endangered tangible cultural items, especially in the particular case of medium and small sized artifacts, we can also cite the deployment of specific storage and protective measures that are intended to safeguard and conserve delicate or particularly chemically susceptible items from the aggressive effects of external elements such as, for instance, excessive light, humidity or heat. The introduction of digital supports and recent technologies issued from the field of computerized automation and processing can certainly provide conservation related activities with new tools to achieve their missions. Among such applications we can mention: the employment of electronic monitoring devices to examine the evolution of the conditions of preservation of sensitive items, in order to allow an adapted and rapid response to their change and the development of computerized numerical simulation to predict the progression of endangering factors that could affect the physical integrity of the concerned items. Likewise, we can also cite the role that three-dimensional computerized visualization technologies can play in successfully assisting the conservative efforts with applications that enable the creation of geometrical 3D representations of cultural items. Such applications can, in fact, be applied for the pre-visualization, simulation, study and preparation of structural modifications and interventions whose objective is the consolidation of the conditions of preservation of the concerned items, as well as for the analysis and virtual testing of possible restoration hypothesis concerning to the prior state of the cultural property during the planning phase that precedes a conservative, or a non-conservative, restitutive intervention Defense Defending and sheltering the items that present a cultural significance against the endangering factors other than the natural aging process, is a vital activity that is necessary to guarantee the long-term preservation and protection of our cultural heritage. In the occurrence of a catastrophic scenario it is essential to ensure that adapted measures are ready to be deployed in order to limit the inevitable consequences that accompany any destructive event, whether caused by human activities or by natural phenomena. Such events can have significantly diverse origins and require the mobilization of a varied array of different actors to guarantee that the physical integrity of the threatened cultural items is preserved and that the subsequent impoverishment of the heritage that bears witness to our cultural identity is prevented. Among the possible natural dangers that could affect the survival of cultural properties we can cite: floods, earthquakes, landslides, fires, excessive humidity and the appearance of natural hostile elements, facilitated by the deployment of insufficient preservation measures, such as parasites or fungi. Likewise, human activities and human interference can as well produce comparable or even superior, degradations and destruction, either deliberate or collateral. Amid such endangering phenomena we can mention: the terrible consequences of armed

54 Conceptualization 53 conflicts, atmospheric pollution, vandalisms, depredations and theft finalized at the contraband of cultural items [PBC04]. Even though the deployment of sufficient conservative and disseminative measures could partially contain or lessen the negative influence that some of those events may exert, the introduction of effective means that are specifically intended to limit the harmful impact that us caused by the emergence of sudden or unexpected threatening situations is however required. In this respect, the intervention and collaboration of organizations such as the police, the fire departments, the civil protection and more generally all field personnel that is directly involved with the active defense of cultural properties against such threats, represents an important factor that participates in the implementation of an effective protection of our cultural heritage in order to guarantee the long-term preservation of its integrity. It is clear that such intervening parties, to efficiently accomplish their mission and to suitably support the expedite decisions that they could be compelled to take on site during critical situations, such as deciding which items to save and shelter and which ones to abandon and disregard, should rely on any available pertinent documentation that has been compiled on the concerned items as they should depend on the knowledge and expertise provided by the investigative efforts in order to correctly evaluate the cultural significance of the items toward which their defending action is directed. The possible actions and measures that can be deployed to ensure the defense of any tangible human cultural heritage, which proceed from the established documentation about the concerned cultural items, can range from the constitution of specific additional security documentations that target all the items that are present at a specific property or site, to the establishment of evacuation plans and intervention scenarios that list all the locations that are occupied by cultural items alongside their pertinent physical attributes and the potential dangers on site that could hinder the salvaging operations. Whenever such interventions should prove necessary, further measures that could be implemented could range from the physical evacuation to a safer place of the concerned mobile items, depending on their importance, to the on site construction of sheltering structures to protect the immobile cultural properties. The introduction of digital tools and visualization related technologies in such areas of activity can assist field personnel with increased capabilities in terms of accuracy, rapidity, reliability and efficiency, during the retrieval of the relevant information that pertains to an endangered item as well as during the planning and deployment of adapted measures that are aimed at the safeguard of its integrity. As a few examples of the contribution that such technologies can offer we can refer to the implementation of remotely accessible multimedia databases to retrieve the information relative to the properties that are affected by an endangering situation and to the creation of three-dimensional representations of cultural items, in order to allow for a fast survey of the features that characterize an intervention site and to provide for an efficient way to spatially localize and recognize all the objects that are susceptible to be evacuated and sheltered. Similarly, we can also mention, as further illustrative cases, the employment of digital acquisition devices in order to constitute the required security documentations and the utilization of computerized simulations to assist

55 54 Methodological approaches for the protection of cultural heritage in the digital age in the preparation of dynamic models that supply early warning capabilities, such as climatic physical numerical simulations as well as meteorological monitoring and analyzing activities Dissemination The activities that participate in the dissemination of the information and knowledge concerning the existing cultural heritage represent a fundamental aspect for the support of the protection and safeguard of cultural properties. In fact, it is only through the diffusion and sharing of such knowledge that the awareness concerning their existence, the understanding of their inalienable cultural significance and their capital importance, as well as the respect that is due for their physical integrity can be promoted, thus limiting the occurrence of deliberate acts of vandalism caused by the pernicious effects of ignorance. The dissemination tasks aim therefore at reaching, communicating and transferring the maximum amount of pertinent information to the largest possible audience, serving an informative and an educating function. In order for such goals to be reached, it is imperative that the acquired knowledge is presented in a suitable format and is communicated via an adapted medium in order to successfully reach the targeted segments of the population for which it is intended. In fact, since most segments of the general population do not usually possess a specialized scientific knowledge of the fields related to cultural heritage, the conveyed information should to be adapted, both in form and content, in order to reach a sufficient degree of accessibility for such audience. To promote the knowledge that is associated with cultural properties, the dissemination activities rely on the available documentation about the concerned items and seek to present the findings and the elements that determine the cultural significance of such items, established by the investigation processes, in a relatively accessible way. To achieve such end, the mobilization and the involvement of several diverse actors is necessary in order to widen the spectrum of possible interventions that assist the dissemination effort. Such activities can range from the creation of television documentaries and printed material, intended to vulgarize specific or specialized knowledge concerning cultural heritage items or sites, to the definition of adapted topics to be presented in the frame of compulsory education, to the reinforcement and facilitation of the generalized access to education and alphabetization, to the organization of expositions, cultural events, to the preparation of commented visits to cultural heritage sites intended for a wide and a varied audience. Among the activities that are relevant for the dissemination of the knowledge linked to tangible human cultural heritage items we can mention the central effort that is provided by museums and, to a certain extent, by private parties in the creation of spaces that host the exhibition of any cultural properties in their possession for the benefit of the general public and the essential role that is played by the educational system in order to spread and transfer the knowledge concerning such items, their history and their inherent significance. Moreover, we can also mention the dissemination activities that are provided by heritage specialists through the edition of written material that is intended for publishing,

56 Conceptualization 55 as well as the physical re-affectation of cultural and historical sites to either their original purposes or to new uses, such as it is the case, for example, for the Arena of Verona. Such measures serve an awareness raising function and can reinforce the role that such cultural items play for their community, since they are constitutive elements of their cultural identity. Similarly, such activities can equally help to lessen the occurrence of potential endangering factors that would follow the abandonment, neglecting and disregarding of such structures: in such respect, the regular use accompanied by an enhanced visibility of such cultural items can trigger conservative interventions to ensure the long-term durability of those properties. The introduction of digital tools in the sphere of content dissemination, and more particularly the application of three-dimensional visualization related technologies, significantly broadens the spectrum of actions that can be implemented for information spreading purposes. Such technologies can support the widening of the capabilities featured by existing analogue traditional methods in order to efficiently transfer the knowledge concerning cultural heritage items and to assist in reaching larger audiences, as well as they can improve the degree of general accessibility to such information. Amid the possible applications that feature a visualization component and that can support the dissemination efforts concerning cultural heritage properties we can cite, for instance, the distribution of remotely accessible electronic catalogues, the constitution of virtual museums and the publishing of culturally oriented educational software and videogames. As further examples, we can also mention the creation of virtual restitutions, in the form of short movies or interactive experiences, the establishment of networked environments for the diffusion and presentation of informative material and the on site deployment of electronic interactive guides, in the form of fixed informative stations, portable devices or augmented reality systems Digitization Alongside the previously introduced intervention axes we can further mention the contribution that is provided by the digitization processes. These processes refer to the application and the deployment of the activities and the methods that are employed to constitute a digital representation of any given pertinent dimension, or meaningful informative data, which is featured by a considered cultural item. The role of the actors that are active in this field is to support, assist and supplement analogue based methodologies with the introduction of new tools that facilitate, expedite or widen the capacities that are offered by traditional approaches through the exploitation of the capabilities provided by the introduction of digital supports and computerized systems. Such activities range from the digital transposition and adaptation of traditional methodological pipelines to a computerized environment, to the creation of new innovative approaches and applications that take advantage of the most recent technologies. Their responsibility encompasses both the digitization of existing information, which is apprehended through the use of traditional tools, and the creation of digitally born data that carries an added value [DIGICULT02]. As introduced by figure 2, these tools can provide supplementary information in the

57 56 Methodological approaches for the protection of cultural heritage in the digital age context of cultural heritage item protection, and can offer new possibilities to explore, investigate and disseminate the content that is linked to additional complex dimensions that pertain to such cultural properties and concerning dimensions that were previously not directly accessible, observable, or easy to capture through analogue based methodologies. Protection of the Cultural Heritage Documentation Conservation+ Defense Investigation Dissemination - Provide access - Provide objects - Provide expertise digital - Create information - Guarantee preservation + protection - Provide knowledge digital analogue Context analogue Exposition Manipulation Interaction Accessibility analogue digital Digitization Analysis Exploration New dimensions analogue Catalogues digital Databases Physical interventions Restoration Conservative measures Recording Digital representation Advise and counsel Figure 2: digital tools as factor bringing a plus value to heritage protection related activities As a final remark, it should be noted that since the deployment of such applications and the interpretation of the subsequently apprehended information, heavily relies on the inherent characteristics of the concerned cultural items, as well as it relies on the knowledge and on the documentation concerning the significance of its distinctive traits, the digitizing mission cannot substitute itself to the other axes of intervention and clearly requires the collaborative efforts and expertise of the fields that it seeks to support in order to provide pertinent and exploitable results that can effectively reinforce and assist the general protection of cultural heritage at all its levels.

58 Conceptualization Groups involved into cultural heritage protection related activities Each of the abovementioned intervention axes and fields of activity that constitute the different facets of the general concept of protection of the cultural heritage, generally engage the predominant responsibility, or the preferred domain of action, of specific groups and organizations. Such groups are usually the firsts that are concerned in the frame of a given activity since they are the more apt to deploy the appropriate measures and tools in order to ensure the protection of a heritage item at its different levels, depending on the extent of the connection that exists between the particular dimensions that are exhibited by the cultural item and their own respective fields of expertise. In the following subsections we will present a categorization of the different parties that are susceptible to be involved into heritage protection related activities, and we will specify the field of intervention, the elements and the requirements that are necessary to allow them to successfully achieve their respective goals. Moreover, such classification will allow us to associate such groups with their respective roles and spheres of activity, which coincide with the application domains of the visualization technologies that we will consider for our study. Furthermore, this classification will provide an overview of the characteristics that identify their distinctive traits: these constitute the defining elements that will be employed for the designation of the different conceptual target end-user groups that we consider the main subjects for the preparation and orientation of the survey and the analysis that will be presented in chapter 4 and chapter Relation between the groups and their intervention areas As we have previously introduced during the description of the different intervention axes that participate in the definition of the general concept of protection of cultural heritage, the degree of effectiveness that such activities heavily rely on the information and data that is provided and exchanged between the actors that are involved in such diverse areas. In this respect, the actors active in the documenting domains provide the necessary information and data that is required by the investigating parties to run pertinent analysis as well as to formulate accurate evaluations of the cultural relevance of the studied items. Subsequently, the combined output that is generated by those spheres of activity constitutes an essential element for the successful implementation of adapted conservation strategies and for the deployment of suitable sheltering defense measures. Likewise, the collected documentation, coupled with the results of the interpretative investigations, supplies the disseminative effort with the required material and necessary data to achieve its goals. In order to summarize such interrelations, as illustrated in figure 3, we ought to consider that some of those actors can be called to play multiple roles: depending on their field of expertise such actors could be involved into several diverse intervention axes at once. It is, in fact, likely that the expertise that is provided by a particular professional category, such as historians, could be relevant for more than one of the activities that participate in the protection of

59 58 Methodological approaches for the protection of cultural heritage in the digital age cultural properties. Consequently, we can regroup a certain number of these actors into a singular category, depending on their specific area of expertise, on the nature of the actions they perform on the cultural items and depending on the degree of vicinity they feature with the cultural properties while accomplishing their tasks. Digital Intermediaries Content Consumers Conservation + Defense Transmission + Dissemination Field Personnel Cultural Intermediaries Investigation + Expertise Documentation + Access Content Holders Cultural Heritage Item Figure 3: relation between the groups active into heritage protection activities To decide in which category these actors are placed, we can also consider the role they play in the generation, transmission or reuse of the information that is shared between the different abovementioned intervention axes, as we can employ the type and nature of the input they require to accomplish their missions, alongside the scope and further utilization of the output they provide as result of their actions. The application of such parameters allows the general classification of the main intervening actors into distinct organized levels that describe their interrelations and their respective contributions to the general preservation of our common cultural heritage. A first category that can be identified contains all the parties that play the role of content holders: the actors that belong to this category are in direct contact with the cultural properties and are responsible for the creation of appropriate documentation concerning the studied items as well as for providing access to such data to the other groups in order to allow them to suitably achieve their goals. Furthermore, such group is equally responsible for the conduction of all necessary investigation activities as for the supplying of the required expertise that is essential to establish a correct evaluation of the cultural significance that is featured by a cultural item.

60 Conceptualization 59 Proceeding from such data, the category associated with field personnel can subsequently provide the application of conservative interventions and the deployment of defense measures in order to ensure long-term preservation conditions for the safeguard and maintaining of the integrity of cultural properties. Likewise, while relying on the results of the investigative efforts and on the previously gathered documentation, cultural intermediaries, which are responsible for the dissemination and transmission of the information and knowledge that associated with cultural heritage items as well as with their cultural significance, can consequently develop effective approaches to both distribute such content, and deploy targeted approaches that aim at reaching larger audiences, and, more generally, aim at establishing a communicative relation with potential content receivers and consumers. Finally, the introduction of new digital technologies that are aimed at assisting the tasks that are accomplished in the frame of the different intervention axes that participate in the definition of the general concept of protection of cultural properties, their implementation and adaptation by information specialists, to suit the respective practical needs that are featured by such diverse fields of application, is entrusted to the mediating role that digital intermediaries are called to exert. Their responsibility therefore includes and encompasses all the efforts that can be deployed at all the levels of interaction between the necessary inputs and the provided outputs that are supplied by the different abovementioned groups, in order to transfer the required information and data from analogue supports to digital ones as well as to create new opportunities and tools intended to facilitate, enhance or extend the capabilities featured by traditional analogue based approaches employed to accomplish their respective missions Content holders The category that includes the actors that play the role of content holders represents the group that works in the closest proximity with the cultural items, whose direct access constitute for them the essential starting point and the fundamental requirement to allow the fulfillment of their primary objectives. The main responsibilities of such group are the ones that are inherent to the investigative and documenting activities. Proceeding from a direct observation, contact and analysis of the cultural properties, they provide exhausting descriptive data, the formulation of analytical studies, which are aimed at increasing the currently existing knowledge of the past, and the elaboration of objective evaluations concerning the cultural significance of the considered items. Moreover, besides offering expertise and documentation, content holders are further responsible for the deployment of appropriate measures that guarantee the access to the information and data they produce in order to assist the efforts of the other intervening groups. Their actions start with the attempt to apprehend and document the nature and significance of the different distinctive traits and dimensions of the cultural items towards which their attention is directed and to which their tools are applied. Their efforts then move from the aspects that pertain to the concerned objects to producing invaluable information

61 60 Methodological approaches for the protection of cultural heritage in the digital age and data that constitutes the essential input for the tasks that are carried out by the other groups that strive to accomplish their missions. As actors belonging to such group, we can cite historians, independently from their specialization fields, who offer interpretative studies, the formulation of pertinent evaluations of the historical value of the concerned items and provide the necessary references to contextualize the cultural significance of the considered properties. As a further contributing professional category, we can also mention archeologists, whose role is to retrieve, excavate and rediscover lost elements that belong to our cultural heritage, and who provide the necessary expertise to carry out comprehensive investigation surveys that assist in the construction of hypothesis that support the extension of the current knowledge and understanding of our past. Finally, as additional actors that participate in such areas, we can cite all the parties that contribute to the documenting task and technically support the investigating efforts, such as, for instance, scientific personnel, who provides specific simulations and precise physical analysis concerning the cultural items for further interpretative studies, data collecting personnel, who is responsible for the gathering of the measurable and observable dimensions that pertain to the studied property and all the actors that are involved in the deployment of effective measures that are aimed at establishing a suitable access to such data for all the other groups that involved into heritage protection activities. It is clear that, as all the other groups that work for the preservation of cultural properties content holders feature specific needs and requirements in order to fulfill their missions: these needs should be carefully considered while preparing the development and deployment of specific tools that are aimed at supporting their work. Since their mission is oriented towards the documentation and investigation of cultural items, the most relevant features that constitute an asset in assisting their efforts are the ones that provide an accurate and precise representation of the dimensions that they seek to apprehend, as well as the ones that enhance their ability to create a context that contributes to the formulation of solid hypothesis and to the better comprehension of the studied items, their past and their historical significance. Moreover, we can further consider all the features that assist them in widening the range of dimensions that they are capable to analyze, thus strengthening their capacity to capture, store, retrieve and transmit the data resulting from the completion of their task Field personnel The field personnel category regroups all the actors whose actions significantly rely on the data and documentation previously produced by content holders, and whose interventions are mainly applied to the physically measurable dimensions and qualities that are featured by the cultural heritage properties concerned by their tasks. The extent of the responsibilities that such group exhibits primarily ranges from conservative interventions to the deployment of defensive and sheltering measures aimed at the preservation and recovery of the structural integrity of such cultural items. The execution of their tasks, which employs the collected documentation and which is supported by the expertise that is offered by content holders to guide and orientate their activities, provides effective intervention and evacuation plans, the establishment of adapted sheltering

62 Conceptualization 61 measures, the deployment of efficient practices to guarantee the safety of such properties in the eventuality of sudden endangering situations that threaten their integrity, and more generally provides activities that enhance their capacity to be preserved indefinitely. Moreover, such group equally implements conservative and restructuring measures, which are aimed at preventing, restoring, reversing or repairing the degradations sustained by cultural items, caused by the impact of the potentially endangering action of slow natural processes, such as aging, by the consequences of the occurrence of external factors, such as pollution, large natural catastrophes or by deliberate destruction, vandalism and theft. Among the professional categories that can be included into this class we can cite the artisans that work in the field of restoration, the personnel that is assigned to restructuring activities and all the groups that seek to preserve and defend the cultural properties from the different threatening elements that could affect their integrity; among these groups we can mention the civil protection, the police, the fire departments and all the professional categories that could be called to exert their expertise to this end. In this respect, we should mention that most of the actors that belong to such category often present an important trait: the professional skills and the particular aptitudes they deploy during a field intervention is usually applied to other types of items in more general ways, and is not specifically connected with cultural properties. A plumber or a mason, for instance, could be called to intervene and employ the tools and expertise that is associated with its profession to consolidate or renovate specific elements, such as a crumbling wall or a defective isolation that is causing water infiltrations into a structure presenting a cultural significance. Such actions are technically the same of those that they could have been executed on a normal structure, but in this particular case, their direct intervention affects a cultural property. Whenever proper guidance is not provided in the occurrence of such eventualities, it is likely that, in most of the cases, the intended work would be executed and carried out exactly as it would have been on any other structure, without considering the particular precautions that should have been implemented in order not to cause any further degradation to the cultural substance of the item during such restructuring interventions. It is therefore essential for field personnel to have the capacity to establish a direct contact with content holders, in order to avail themselves of their assistance and expertise in identifying and properly considering the relevant cultural elements that should be targeted by their interventions while avoiding the occurrence of any potential collateral damage caused by their actions. Furthermore, it is vital that on site intervening parties equally benefit from a practical and transparent access to all the pertinent information and data that has been produced by content holders, both in order to better plan the execution of their interventions and to identify the specific cultural items that are the more susceptible, for their significance, to be the target of their efforts. The different tools that can assist field personnel in the execution of their tasks and that could allow them to suitably perform their duties should mainly provide them with accessible methods to retrieve all the pertinent documentation they require concerning the cultural items that are targeted by their activities; these tools should equally offer the capacity to strengthen and support the establishment of efficient coordination, exchange and direct communication

63 62 Methodological approaches for the protection of cultural heritage in the digital age practices with cultural holders to better guide and support their interventions. Moreover, such tools should as well exhibit the ability to properly assist all intervening actors by providing them with precise information and reliable data that could help them determine the significance of the endangered cultural properties, the priorities that are associated to their safeguard, the particular measures that should to be deployed in respect to their state of conservation and the precautions that should be considered when storing or moving such items. More generally, in order to guarantee the optimal deployment of effective and adapted measures aimed at reinforcing the impact of conservative and defensive interventions as well as the integrity of the threatened cultural properties, these tools should provide an increased efficiency, in terms of facilitating, assisting and expediting the execution of their tasks in situations that might require an expedite intervention due to the presence of time constraints imposed by external factors, such as a fire or a flood Cultural intermediaries The category that contains cultural intermediaries encompasses all the actors whose actions are based on the collected documentation and on the data that is produced by cultural holders, and which aim at establishing a bridge between the cultural items, their related cultural aspects and external groups that do not necessarily pertain to those categories that are directly connected to heritage protection activities. Promoting external access to such information can encourage the diffusion of the knowledge about cultural heritage as a whole and therefore can assist in limiting the occurrence of deliberate destructions of cultural properties that are facilitated by ignorance related issues. The responsibilities that cultural intermediaries face are primarily related to the dissemination and transmission of the knowledge, the documentation and the information that is connected with cultural heritage properties, and to the promotion of the comprehension and understanding of the peculiar aspects that constitute the cultural significance of the items that bear witness to our cultural identity. Secondarily, some cultural intermediaries, such as museums for instance, support the preservative and sheltering efforts by providing the physical space necessary to store, conserve and exhibit the cultural items while they attempt to establish a contact between the cultural properties and the groups that are targeted by their divulgative efforts. While accomplishing their tasks, cultural intermediaries establish multidirectional communicative channels and relations that facilitate the transfer of heritage related information to external actors and they create effective transmission opportunities that ensure the diffusion of the such knowledge to its intended audience through the exploitation of appropriate medias that raise the general awareness towards the different issues and aspects linked to the preservation and protection of the cultural heritage. Furthermore, cultural intermediaries also regulate the access to the cultural properties while they supply the documentation and context dependant information that is required by external groups to fully comprehend the cultural significance of the concerned items, their relevance for our common cultural heritage, their interconnections with our past history and their contribution to the

64 Conceptualization 63 constitution of our cultural identity: as such, they ultimately play an educating and instructing role. Amid the different organizations and professional categories that are contained in this class of actors we can mention: divulgative agents, such as journalists, editors or audio-visual media providers, which supply the means to support the diffusion and publishing of articles, books or documentaries concerning cultural heritage related issues and all the researchers and groups that produce the media content that is required for such disseminative activities. Furthermore, we can cite the essential task that all the actors that are active into educative structures play in respect to the transmission to large portions of the population of both the knowledge relative to cultural properties and the conceptual tools necessary for its comprehension, hence reinforcing the preventive efforts that are aimed at reducing the occurrence of deliberate and collateral degradations which affect cultural items as a result of ignorance about their significance and uniqueness. Finally, we can mention the central role that is played by museums with the organization of activities that are aimed at providing the public with the broadest and fairest access possible to the cultural items they conserve, and with the creation of spaces that feature suitable storing conditions in order to preserve and shelter delicate cultural items from the impact that external aggressing factors could exert on their structural integrity. In this respect, it has to be noted that such institutions could also serve, to a certain extent, as a nexus that could support, assist, coordinate and concentrate the actions and the efforts that are carried out by the other groups involved into cultural heritage related activities. Thus, beside their inherent divulgative and disseminative vocation, museums can equally assist all the groups that are active into the fields of conservation and preservation as they could support the investigative and documenting efforts by providing all the actors that linked to such activities both with the access to their collections and with their extensive expertise and experience they posses concerning specific cultural items. The features exhibited by the tools that are susceptible to be employed for the dissemination and transmission tasks that could positively affect the outcome of the different activities that are carried out by cultural intermediaries are generally related to the aspects that can strengthen their capacity to reach larger audiences or allow the information that they intend to communicate to be presented in clearer or more accessible ways. Such features should provide disseminative agents with more effective ways to transmit complex information and with simple and effective approaches that enhance the clarity of their messages while facilitating their understanding by non specialists. Likewise, they should equally offer the possibility establish a communicative relation with a wide range of potential receivers in order to guarantee a maximum coverage, a large diffusion and an optimal impact for the execution of their intended divulgative efforts. Furthermore we would like to note that, in order to ease such tasks, the deployment of specific tools that allow, either passively or interactively, the display of complex data alongside its associated informative content by employing straightforwardly accessible visual references as supports or

65 64 Methodological approaches for the protection of cultural heritage in the digital age interfaces, can constitute a valid asset to enhance the visibility and the effectiveness of the disseminative tasks, as well as it can encourage the emergence of an active interest from the targeted audiences towards such content Content consumers The actors that usually do not assume any active role in either of the activities that are directly associated with the concept of cultural heritage protection, and who constitute the intended target for the disseminative and divulgative efforts that are deployed by cultural intermediaries, are regrouped in the category that encompasses all the parties that are susceptible to act as content consumers. Such group usually does not feature a high level of direct proximity or a high degree of direct physical interaction with the cultural items; however, this group tends to reflexively apprehend the dimensions that characterize the cultural importance of such items through an indirect and guided relation that is mediated by cultural intermediaries. Furthermore, all the actors that belong to this group usually do not exhibit the sufficient degree of expertise and technical competence that is required to comprehend in detail all the implications and the meaning of the raw information that is produced by the investigating parties and that is conveyed by the disseminative agents. Thus, it is essential that such content is presented and diffused with a suitable degree of accessibility and that is accompanied with the necessary explanatory notes in order not to discourage the receiving parties to engage themselves in the process that will lead them to a full understanding of such aspects. Content consumers therefore represent the target for the divulgative efforts and in such capacity they are the intended audience for the transmission, adaptation and dissemination of the information that is provided by content holders for the benefit of non specialists. While it is necessary for any communicative practice to display a sufficient degree of mutual collaboration between the groups that are engaged in such processes in order to guarantee a successful transmission of the originally intended meaning, the relation between cultural intermediaries and content consumers is however unidirectional due to the technical nature of the raw information and data that is susceptible to be disseminated. It is therefore essential to properly consider the degree of adaptation that has to be applied to content to be disseminated as to the type of medium that is employed to convey its associated meaning in order to maximize the chances that the intended audiences are reached and that the effectiveness of the divulgative effort is ensured. In this respect, it should be noted that the presence of some sort of feedback mechanisms between cultural intermediaries and content consumers is desirable in order to confirm the suitability of the media that is employed to convey the intended messages, the extent of the impact of their communicative efforts and the amount of interest that the public has exhibited toward such actions. Monitoring such aspects could therefore prove useful to correct and revise any failed disseminative strategy as well as to adapt existing strategies to better suit the expectations that characterize their targeted groups: such

66 Conceptualization 65 monitoring should then provide the necessary information to enhance the overall efficiency of those strategies and would strengthen their ability to effectively reach larger audiences. In view of these considerations we can regard the needs of content consumers as a complementary extension and a contributing element that participates in the definition of the features that the tools employed by cultural intermediaries should exhibit in order to facilitate the accomplishment of their tasks. In this respect, amid the significant distinctive traits that can positively affect the response of content consumers when confronted with the disseminative efforts that are initiated by cultural intermediaries, we can name all the features that would reinforce the overall clarity and the overall accessibility of the transmitted message as well as those that would facilitate the comprehension of the conveyed knowledge through the adaptation of such content for a non specialist audience. Furthermore, we can equally cite all the features that can participate in the definition of content transmission strategies that are capable to both increase the interest of content consumers toward cultural items and provide the capacity to create appealing contexts that are susceptible to encourage the targeted audiences to actively seek access to the already available knowledge, therefore creating the necessary conditions to guarantee their receptiveness. Finally, we can also mention all the aspects that can contribute to supply content consumers with practical and accessible opportunities, tools and media that would allow them to easily discover, apprehend and understand the content and the knowledge that is associated with cultural heritage properties Digital intermediaries Digital intermediaries regroup all the actors that are involved in any digitization activity that is related to cultural heritage properties, and whose tasks range from assisting cultural heritage preservation oriented efforts to achieve their objectives, to supporting all the actors in such fields though the development of new methodologies and digital applications that can facilitate their tasks. Their responsibilities vary from the identification of new digital based approaches that meet the specific needs of these abovementioned actors, to the implementation of new technologies and tools that feature a suitable degree of adequacy to the practical requirements that are exhibited by the groups they seek to assist. In such context, digital intermediaries assume a mediating role between the technical knowledge they posses about the possibilities that are offered by the new technologies of information, communication, and visualization and their potential application to existing fields of heritage protection that posses their own established methodological pipelines. However, their capacity to perform such supporting task is not exclusively confined to restricted collaborations with a limited number of actors, but it extends to all the interactions that exist between all the previously identified groups active into heritage protection related activities. In this respect, they contribute to the definition of new methodologies that specifically enhance the capacity of other groups to accomplish their missions, as they provide new and existing tools, through the deployment of

67 66 Methodological approaches for the protection of cultural heritage in the digital age their technical expertise, with new features and functionalities that increase their overall efficiency. While all the groups that play an active role in cultural heritage protection usually rely on a defined set of inputs for the execution of their activities, and generally present a relative autonomy in the production of the results they deliver, digital intermediaries, on the other hand, need to consider, due to their particular mediating function, the combined needs and requirements displayed by all the groups that participate in the interactions they seek to enhance. Digital intermediaries, by promoting the implementation of technologically oriented solutions that are aimed at modernizing the approaches employed for the protection of cultural heritage, play a role that requires strong multidisciplinary aptitudes in order to comprehend the issues that are related to the different fields in which they intervene: such task presents a noticeable interdependence with the activities performed by all the groups that are concerned by the potential introduction of such tools. Consequently, in order to produce any useful application and offer any positive practical and methodological contribution, it is essential that digital intermediaries closely rely on the feedback, on the knowledge, on the guidance and on the advice provided by technical experts and other cultural heritage actors. Such practices, interactions and exchanges would ensure that the developed digital tools fully consider the respective needs of those groups as they would guarantee a suitable degree of adaptation with their requirements: such collaboration would then represent a valuable asset for any group that is involved into heritage protection activities. Amid the professional categories that we consider digital intermediaries we mention all the actors that apply the technical expertise they possess in digital domains to assist, either directly or indirectly, the protection of cultural heritage. As such, we can cite: theoretical researchers, application oriented developers and technical manufacturers, either active into hardware or software related activities, who work with cultural actors while directly supporting their efforts and providing them with new digitally based applications and tools which offer to those groups new possibilities for the accomplishing of their tasks. Subsequently, we can also mention all the groups that participate in the practical deployment of such technologies on the field, and whose role is to create, maintain and implement all the supporting media that is necessary to allow their correct use in concrete situations. Finally, we would like to cite as further members that can be assigned to this category: technical personnel, information specialists and all the personnel that posses the capacity to create and manipulate digitized content, either digitally acquired or digitally born, such as for instance multimedia and visualization experts. In view of the role that digital intermediaries are required to play, we can identify the elements that can simplify and facilitate their mediating task. In this respect, it is clear that such aspects are the ones that would enhance the capacity to understand the practical and theoretical issues that are associated with the spheres of interventions in which cultural agents are active, or the ones that would assist them in clearly apprehending the specific needs and priorities of all other cultural agents. Amid those elements we can certainly mention the ones

68 Conceptualization 67 that strengthen and encourage the instauration of exchange, consultation, cooperation and feedback practices that should exist between digital intermediaries and all the groups toward which their supporting actions are directed: the instauration of such practices is essential to allow the rapid development of adapted applications that guarantee an increased usefulness for any newly developed digitally based approach. 2.5 Summary In this chapter, a first initial definition of the concept of cultural heritage property, which has been derived from The Hague s convention of 1954, has been introduced and further expanded, due to its originally limited application range, to encompass a larger spectrum of cultural items to which such definition should be applied. In order to apprehend the extension of such application range, the identification of additional pertinent dimensions, which could be relevant for the characterization of any potential cultural property, has been undertaken. Hence, distinctive traits such as size, inherent nature, tangibility and cultural significance have been employed to constitute a general classification of such properties according to their exhibited characteristics. Furthermore, through the exploitation of such categorization, the scope and focus of our work, in relation to the type of heritage item we are primarily concerned with, and towards which our attention and interest is principally directed in the frame of this study, has been defined as concerning tangible human cultural heritage. Subsequently, the general concept of cultural heritage protection has been discussed and the different interdependent aspects and intervention axes that participate in its definition and application, and whose collaborative efforts and joined contribution represent an essential factor to ensure the effectiveness of such enterprise, have been determined and clarified. Additionally, the activities that have been identified as related to the concept of heritage protection, such as documentation, investigation, conservation, defense and dissemination, alongside the digitization processes aimed at providing such tasks with new methodological approaches and tools that assist and reinforce their efforts to preserve our common cultural heritage, have been analyzed and examined along with their specificities. Moreover, amid the different digital tools linked to the newly available information and communication technologies that that can be applied to the different areas of intervention connected to the protection of cultural heritage, the ones that are issued from the fields of computer based visualization, threedimensional representation and interactive exploration have been highlighted as central focus for our work. Finally, all the groups and actors involved or participating, whether actively or passively, in any of the axes that reinforce the general protection of cultural heritage, have been identified and classified alongside the relations they feature with all the other intervening parties, the particular needs they display and the aspects of their tasks that can benefit the most from the introduction of new digital tools specifically aimed at easing certain facets of their missions.

69 68 Methodological approaches for the protection of cultural heritage in the digital age Thus, content holders, field personnel, cultural intermediaries, content consumers and digital intermediaries have been defined as general conceptual containers to categorize and regroup such actors, depending on the degree of vicinity they exhibit with the cultural items concerned by their interventions and depending on the nature of the role they are required to play in the frame of the preservation and protection of the properties that belong to the human tangible cultural heritage.

70 Related work: taxonomy and review 69 Chapter 3. Related work: taxonomy and review In this chapter, the different approaches that can be employed to constitute a visual representation of a cultural heritage property in the frame of any of the previously activities related to the protection of our cultural heritage, will be presented and analyzed. Thus, a general classification of such approaches, which range from the ones that feature the use of traditional tools to the ones that are based on modern digital technologies and computer aided visualization techniques, will be constituted. In order to establish a coherent taxonomy, specific dimensions and axes that characterize such methods will be discussed and further employed to define a general conceptual framework that will allow the classification of all possible design choices according to their specific characteristics. Furthermore, for each combination of the identified axes, a few selected examples will be proposed and specifically discussed to highlight the respective advantages and disadvantages of the considered category of methods. Finally, a selection of approaches and modeling techniques which are specifically related to the virtual representation of heritage items by means of modern visualization technologies will be explored and discussed. Then, a subset of the previously defined dimensions will be employed to present and classify such methodological approaches while emphasizing their specific strengths, weaknesses and practical usefulness in relation to the activities that are related to the protection such cultural heritage properties. 3.1 Methodological axes and dimensions The main objective of this chapter is to propose a coherent overview of the methodological approaches that are available to constitute a visual representation of a heritage item. However, offering an extensive and comprehensive analysis of all the techniques and technologies that have been used in the past years in the frame of several different projects each aimed at achieving such goal, would prove impractical due to the amount of combinations and the diversity in approaches that have been employed. To overcome such problem, a general classification that is based on specific characteristics that can be found and identified at varying intensity across all such projects is proposed instead. As such, even though only a few selected examples are offered to illustrate the specificities of each combination of the chosen traits, which are employed to define a set of specific categories, it is still possible to fit by analogy other approaches, that were here omitted for practical reasons, by employing the proposed conceptual canvas. To establish a clear and generic classification that can encompass all the design choices and methodological approaches that can be used to visualize a heritage item, by employing either traditional or digital techniques, a minimum set of dimensions is desirable to ensure a sufficient flexibility of our conceptual model. Consequently, to assist the creation of a coherent taxonomy, five continuous dimensions, or continua, [MILGRAM99] [MILGRAM99b] [AZUMA97] have been retained. These continua will be employed as axes to define a conceptual space which allows the classification of the selected approaches, thus assisting in the

71 70 Methodological approaches for the protection of cultural heritage in the digital age highlighting of their relative differences, similarities, limitations and qualities depending on the dimensions that most influence their placement in the defined taxonomy Continuum of virtuality Since we aim at classifying the visualization attempts which are targeting heritage items and which are employing indistinctively either traditional or digital approaches, the definition of a continuum that accounts for the amount of virtuality present in a given visualization is a necessary dimension to be considered. Following the approach proposed in [MILGRAM94] we maintain that, although real and virtual environments may be considered as two distinct and separate entities, they do not however offer a mutually exclusive alternative and they can be combined and mixed to a varying degree thus defining a continuous range of intermediate choices. Any location along the continuum of virtuality, as is presented in [MILGRAM99] [MILGRAM99b], implies, for a computer, the possession of a corresponding degree of digital knowledge about the represented real world. In other words, a maximum degree of virtuality implies for a computer to possess complete quantitative information about the represented environment, which therefore needs to be entirely modeled in order to be rendered. On the other hand, for a totally real environment, which is completely unmodeled, a computer does not possess any information about the content nor the meaning of the visualized properties, including their relative position, their significance, or any other information pertaining to such items. Consequently, the definition of real environment can equally encompass not only live experiences, but can be extended to include raw photographic data, scale models, scanned architectural plans and restitution drawings which, even though they may per-se present some elements of virtuality, in the sense that some drawn elements may have been added to the real data, as per the presented definition they can nevertheless be located on the reality portion of the continuum Continuum of interactivity A second dimension that needs to be considered to classify the considered approaches is the one that accounts for the varying degree of interactivity which is exhibited by a visual representation of a heritage item. Interactivity, as conceptual construct, has been widely studied [RAFAELI88] [LIU02] [MCMILLAN02] [SVANAES00] and several definitions have been developed focusing either on user-user interaction [DEIGHTON96], user-machine interaction [STEUER92] [RAFAELI07] or user-message interaction [CHO97]. However, for our needs, a definition that specifically accounts for the relation between the user, the visualized items, either virtual or real, and their interaction capabilities is necessary to establish an axis suited for our classification. In [SVANAES00] an interaction is defined as an action involving at least two participants: in the context of a human-computer interaction, the human is interacting with the computer. For an item to be considered interactive it is required that it allows for a certain degree of meaningful interaction. The term

72 Related work: taxonomy and review 71 interactivity hence denotes the interactive aspects of such item. Therefore, we consider the degree of interactivity featured by a given artifact, a specific item or a particular situation, as its capacity to contextually offer the human user the possibility to subjectively experience an interactive behavior in a synchronous way, thus enabling the user the opportunity to either actively contribute to a given experience or to affect, modify or influence in real time, either directly or indirectly, the considered item. Along such continuum it is consequently possible to distinguish between highly interactive approaches, which allow the user to actively participate and manipulate the visualized items, and static approaches in which the users are presented the items to be passively observed without any possibility to change, to modify, to direct nor influence their subjective experience Continuum of precision A further multifaceted dimension that should be considered is the continuum of precision. With the term precision, we do not exclusively, nor predominantly, refer to the amount of geometrical detail and data that can be fit into a computer aided visualization, a scale model or an architectural plan, although such element represents an important component of its definition, but we extend such concept to encompass all the elements that can contribute to enhance the consistency, both historical and scientific, of a given visualization attempt. Therefore, information related for instance to sociological or geographical data, such as specific activities that were taking place in a certain place in the past, the practical every-day use for which a specific artifact was employed, the type of garments that were common at a specific time, the language that was spoken or the climate that was prevalent in that region, whenever present and correctly considered, would be as equally important as having a high geometrical fidelity. The presence of such dimensions would then allow the user to better contextualize the visualization attempt and raise its overall global precision, intended here as the degree of proximity that is exhibited by the visualization in respect to the physical, historical, sociological and geographical realities of the considered item. Consequently, for instance, a live experiment that recreates a Greek battle in its original site would not be as precise as it could be if elements as the type of weather that was present at that time, the time of day the battle began or the weight of the armors that were employed were not fully considered. Similarly, a computer aided visualization of an ancient Roman city, even though not featuring an extreme degree of geometrical detail and accuracy in the reconstruction of the edifices, but which would illustrate, by means of the inclusion, for instance, of narrative elements [ANSTEY00] [NANDI00] alongside the visual ones, the life of its inhabitants, the use they made of their tools and the practices that were commonplace at specific places or buildings, might prove to be a more precise reconstruction than a very detailed wooden scale model of the same city Continuum of visual consistency In Computer Graphics and in traditional arts, more specifically in painting, photorealism can be described as the quality possessed by an artificially created

73 72 Methodological approaches for the protection of cultural heritage in the digital age image to be indistinguishable from a snapshot of the real subject it depicts, and thus that could be easily mistaken for a photograph. In the specific case constituted by our attempt at categorizing the traditional and digital approaches aimed at the visualization of heritage items, such definition would probably generate some confusion, since one could argue that reality is per-se tautologically photorealistic. Nonetheless, we could also consider that a partial reconstruction on site, based on the remains of an ancient edifice and built on top of them by employing modern construction materials, would prove rather unfitting and would not account for the past reality of the site, hence finally ending up to exhibit a lesser degree of photorealism, in respect to a theoretical snapshot of the structure that might have been taken in the past, than it could have been expected a priori. Therefore, to simplify and expand the fundamental concept illustrated by the previous example, we introduce the concept of visual consistency as a further continuum to be used in our characterization. Visual consistency is here intended as the capacity for a visualization attempt, either digital or created by means of traditional methods, to stimulate in the user the subjective visual impression that the considered items would be indistinguishable from reality even if they were placed in their corresponding spatial and temporal original context, thus allowing the establishment of a suspension of the viewer s disbelief [HOLTZMAN98] [SHAPER78] in front of the proposed visualization. In other words, it is the capacity to provide the users with the subjective impression that even if they were placed in the past to directly observe the reality of the considered item at a specific time, they would not notice any discrepancy between such past real snapshot and the proposed visualization; likewise, it could also be described as the capacity to create a coincidence between the visual qualities of the proposed visualization attempt and the mental image that the user has of the visual qualities that the represented item should have possessed in the past. As a side note, it is important to stress that geometrical precision does not necessarily coincide with visual consistency: both qualities may of course coexist in the same visualization, and often this could be desirable; however, the presence of a high degree of precision does not automatically imply the presence of a high degree of visual consistency. In fact, we could easily imagine a 3D model of a given edifice, such as, for instance, the coliseum of Rome or even a fictional building, which exhibits an extremely high degree of visual consistency without offering at the same time a rigorous representation and reproduction of all the details its real counterpart presents. Conversely, we could also consider an extremely detailed replica of a common animal or an extinguished predator, such as the ones that can be found in museums of natural history, or we could also consider a complex and detailed plaster scale model of the reconstruction of an ancient building: in all these cases, even though their degree of precision might be extremely high, their visual consistency is not in par with the visual qualities that are, or were, exhibited by their real counterparts Continuum of automatism and span of development cycle The fifth, and last, dimension to be considered is the one constituted by the degree of automatism, or span of the development cycle necessary to complete

74 Related work: taxonomy and review 73 a visualization project, which accounts for the effort necessary for the acquisition and processing of the data required to produce a specific visualization attempt. In other words, such continuum represents the total amount of time, in terms of manual direct intervention, research, design, post-processing, development and modification, that is necessary to invest in order to prepare and finalize, though the use of either traditional or digital techniques, a specific visualization project which is targeting a cultural heritage property. Along such continuum it is therefore possible to identify and distinguish between the approaches that necessitate a minimal, or modest, amount of external intervention to produce a usable result, and the ones that inversely require the effort of several people, demand the investment of a conspicuous amount of time or involve a painful process of redesign whenever some modification to the final result would prove necessary in order to meet the goals of the visualization attempt. Amid the approaches that are positioned on the far end of such continuum, we can cite, for instance, the execution of photographic documentations, the creation of geometrical models through laser scanning techniques and photogrammetric reconstructions. Conversely, among the methods that sit on the low end of the automatism continuum, we can mention the manual creation of complex 3D models, the manufacturing of precise wooden scale models or the preparation and post processing of complex and historically consistent movies Typological classification models for the existing approaches Having first defined the continuous dimensions that constitute the main axes upon which is based our attempt at categorizing all the existing approaches that can be employed for the visualization of heritage items, namely the degrees of virtuality, interactivity, precision, visual consistency and automatism, it is now necessary to combine them to establish a conceptual framework which is capable to encompass and classify all such approaches. To assist the definition of such conceptual model and ensure its coherence, it is equally essential to explore whether or not some relations of proportionality, either direct or inverse, exist between these continua. As a first step towards such goal, it is important to consider if it would be possible to conceive a specific use-case which would be characterized by a simultaneous maximization of the intensities of all the previously defined axes. Since the first continuum defines the degree of virtuality exhibited by a given approach, we can safely assume that the variation of the intensity featured by such dimension would not exhibit a significant direct impact or link with the variation of the other ones. In fact, such dimension is mainly employed to differentiate between that visualization projects that rely on traditional approaches and the ones that otherwise exhibit a preponderant implementation and use of information technology related tools. Thus, our attention should be directed towards the relations that might exist among the other continua. As it was previously introduced, the maximization of visual consistency and precision are not mutually exclusive and can coexist while being implemented jointly in the frame of the same visualization project.

75 74 Methodological approaches for the protection of cultural heritage in the digital age However, as it was equally illustrated by the short examples given in section 3.1.4, both dimensions can be observed at their maximum intensity both separately and independently, without involving, nor requiring, the simultaneous presence of the other. Moreover, in either configuration, both dimensions can be maximized in the frame of the same visualization project alongside the continuum of interactivity without displaying any direct link of causality with it. It is in fact possible to consider a situation, such as a 3D photogrammetric reconstruction, in which the interactivity of the visualization is ensured, due to its geometrical simplicity, and which exhibits a high degree of visual consistency without implying a high degree of precision. Conversely, a precise scale model might feature a high degree of interactivity without necessarily implying a high visual consistency. A few examples for the possible combinations that can be found between visual consistency, precision and interactivity can be found in Table 1 to further illustrate the mutual independence of such dimensions in respect to each other. interactivity visual consistency precision example high high high complex real time simulation high high low photogrammetric modeling high low high interactive scale model low high high laser scanned 3D data high low low simplified 3D real time model low low high scanned raw data low high low photographic documentation low low low manually drawn sketch Table 1: combinations between visual consistency, precision and interactivity Finally, it necessary to consider the relation that exists between such dimensions and the continuum of automatism. Since a high degree of automatism implies a shorter development cycle for the completion of a given visualization project, it is intuitively obvious that being able to finalize a visualization attempt which simultaneously features a good precision, elevate visual consistency and a high interactivity in a short amount of time is rather unlikely. Since the possibility to maximize all of these continua at the same time is not a realistic option, it is therefore necessary to evaluate the trade-off that exists between these dimensions, as illustrated by Table 2, and to consider that a high degree of automatism, i.e. a short development cycle, may exclusively coexist, in the frame of the same visualization project, either alongside a high interactivity, or alongside either a high visual consistency, a high precision or both.

76 Related work: taxonomy and review 75 development interactivity visual consistency precision example long high high high complex real time simulation long high high low photogrammetric modeling long high low high interactive scale model short low high high laser scanned 3D data short high low low simplified 3D real time model short low low high scanned raw data short low high low photographic documentation short low low low manually drawn sketch Table 2: combinations between visual consistency, precision, interactivity with their corresponding shortest development cycle While choosing a specific approach for a visualization attempt, it is essential to correctly judge whether or not the expected results ought to be produced in a given time frame, which elements are the most important to be implemented, and which dimensions should be prioritized over the others in the trade-off between visual consistency, precision, interactivity and development cycle. As a side note, it is important to stress that Table 2 specifically presents only cases in which, for each combination of the degrees of interactivity, visual consistency and precision, it was possible to identify an example that was featuring the shortest possible development cycle. Therefore, it is generally feasible, for each considered combination whose development cycle is noted as short, to propose a second example that is exhibiting similar qualities but that is featuring a lesser degree of automatism. For instance, if we consider laser scanned data, which is noted as having a short development cycle, a low interactivity, a good visual consistency and a good precision, we could propose the example of CG movies as a counterpart that is exhibiting the same qualities while generally requiring a longer development cycle. Visual Consistency and Precision Interactivity Automatism Figure 4: conceptual space defined by the relation between the continua of visual consistency, precision, interactivity and automatism

77 76 Methodological approaches for the protection of cultural heritage in the digital age To illustrate the connections that have been identified so far between such four continua, which can assist the definition of a conceptual space suited for the classification of the approaches used for the visualization of heritage items, we chose to employ a triangular shape, as illustrated by figure 4. Each vertex of the conceptual triangle acts as an attracting pole and represents one of the continua previously discussed, with the exception of the continuum of reality that will be specifically implemented in a second, more generic, conceptual model. The continua of visual consistency and precision have been regrouped in the same region and are sharing the same spot on the conceptual triangle. This choice has been made to account for the following considerations. Firstly, as previously discussed, such continua can coexist, alone and together, with either of the dimensions that occupy an adjacent vertex. Secondly, one of the main attributes that should always be exhibited by a valid and coherent visualization of a heritage item is a global precision as pronounced as possible which should be limited only by the constraints inherent to the chosen approach, since an imprecise representation would impact on the scientific reliability of the results. Lastly, in most of the cases, once a certain degree of precision is attained in the frame of a rigorously prepared and scientifically sound cultural heritage visualization attempt, it is both desirable and often achievable, whenever allowed by the chosen approach and by the specific trade-offs it implies, to maximize its visual consistency, although that can generally be done at the expense of some degree of automatism. Any considered approach can therefore be compared while being classified as a single point located inside the defined conceptual space: the degree of vicinity of such point to a certain vertex, or the intensity of the attraction exerted by a specific pole on the considered point, would account for the capacity of the corresponding approach to maximize a specific dimension at the expense of the others. Since it is possible to simultaneously obtain the maximization of only two of the three dimensions in the frame of a given project, the classified approaches will generally be located either alongside its segments or in the neighboring area. As such, if a given approach is located around the midpoint of a segment, it would indicate that both dimensions that reside on the opposed ends of the considered segment are simultaneously present with a similarly high degree of intensity in the frame if a given visualization at the full expense of the third conceptual pole. Expressed in a simpler manner, the shorter is the distance that separates the location of a point from a vertex, the more affinity the considered approach exhibits towards the continuum that is represented by such conceptual pole. Consequently, a point that is located inside the taxonomical space will feature shorter distances from its main exhibited dimensions than the one that separates its position from the third conceptual pole, which constitutes the primary tradedoff dimension. Therefore, the more a point is located near the edges of the conceptual space, the more pronounced is the trade-off featured by the considered visualization strategy in respect to its third attribute. As a result, it is unlikely to encounter an approach being classified at the exact center of the conceptual model, since it would imply that the considered visualization attempt would simultaneously feature a total absence of intensity alongside all the

78 Related work: taxonomy and review 77 continua represented by the three attracting poles of the taxonomical space, and that all the continua would hence progress in the same direction. Interactivity Virtuality Figure 5: conceptual space defined by the relation between the continua of interactivity and virtuality The previously presented taxonomical model, however, does not consider as a defining dimension the varying degree of virtuality featured by a given visualization. Although being sufficiently flexible and generic to allow the encompassing of both traditional and digital approaches, such conceptual construct has been nonetheless initially intended as a tool to classify and analyze the respective differences, strengths and weaknesses specifically exhibited by computer aided modeling approaches, virtual representation strategies, and computer assisted visualization technologies, which constitute the central focus of our work. In order to include such dimension, and thus extend the range of cases susceptible to fit in the presented taxonomy, it is necessary to define a second conceptual space, introduced in figure 5, which specifically considers the virtuality continuum as one of its a main axes. In order to build a practical taxonomical model, which can be used to extend the range of our conceptual tools, it is essential to identify at least a second dimension to be employed for the construction of such classification space. Since the relations between the other four considered continua have already been used in the previously presented model, and in order to keep the taxonomy as simple and as clear as possible, the choice not to extend the second conceptual space beyond two dimensions has been preferred. However, since we considered only five characterizing dimensions to support the creation of the proposed classification spaces, it is inevitable that whichever continuum we would choose as second dimension to be included in the this new conceptual model, it would already have been employed in the previously defined taxonomical space. The question that has to be answered is therefore which of those four continua represents an appropriate choice to define, alongside the

79 78 Methodological approaches for the protection of cultural heritage in the digital age continuum of virtuality, a two dimensional space that can provide a suitable categorization of the considered approaches. The continuum of precision and the continuum of visual consistency, as previously explored, exhibit coexisting capabilities without featuring neither mutually exclusive nor mutually inclusive properties. Furthermore, they display a flexible relation with the other continua that define the first conceptual space. Thus, their inclusion in the second conceptual model could generate an unnecessary combinatory complexity that would defeat the purpose of simplicity and clarity which are sought for the creation of our general taxonomy. A similar conclusion can also be drawn concerning the inclusion of the continuum of automatism: in fact, its inability to concurrently develop in the same direction as the other three continua that were retained in the first conceptual model, and consequently the subtle trade-off considerations its maximization implies, would present an unnecessary complication for the elaboration of a clear and practical conceptual classification space. For these reasons, the choice to use the continuum of interactivity as second component for the definition of a two dimensional taxonomical conceptual space has therefore, been preferred. This two dimensional model, which allows for a more generic view of the considered approaches than the initially introduced multidimensional one, will be employed in section 3.2 as a tool to perform a first general categorization of the visualization strategies that target cultural heritage properties. The considered approaches will be classified as points located inside the defined conceptual space according to their respectively exhibited degrees of virtuality and interactivity. Subsequently, fulfilling its originally intended purpose, the multidimensional space defined in our initially developed conceptual model will be employed in section 3.3 to formulate a more detailed classification and characterization of the approaches and modeling strategies specifically related to computer aided visualization techniques and digital technologies. 3.2 Overview of the identified approaches In this section we focus our attention on the constitution of a general taxonomy of all the approaches that can be employed for the visualization of a heritage item. The two dimensional conceptual space created by the association of the continuum of virtuality with the continuum of interactivity will be employed to classify such approaches accordingly to the degree of intensity they present in respect to these two specific dimensions. To assist such task, and in order to simplify the presentation of the considered visualization strategies, as well as to allow the definition of general categories to be used for our classification, the taxonomical space has been subdivided into four main quadrants, or families, as depicted by figure 6. Each of these areas represents one of the four possible combinations between the different degrees of intensity that can be exhibited by a specific visualization attempt in respect to the dimensions that are characterizing such approach. Although both dimensions defining the conceptual space are continuous, hence theoretically allowing for an infinite amount combinations and intensities along their axes, the creation of four generic classes that are employed for our taxonomy resides on the adoption of a broad distinction between the

80 Related work: taxonomy and review 79 visualization attempts that preponderantly exhibit a high intensity, in respect to a given dimension, and the approaches the conversely feature a low one. Second quadrant Fourth quadrant Interactivity First quadrant Third quadrant Virtuality Figure 6: definition of four main taxonomical families inside the conceptual space created by the relation between the continuum of virtuality and the continuum of interactivity Such distinction is exclusively employed to classify the visualization strategies relatively to each other, as well as to provide a model to schematically identify to which family such approaches belong. Hence, by placing a visualization strategy inside a given quadrant, it is still possible to visually express in a more accurate fashion the exact degree of virtuality and interactivity exhibited by a specific approach. In order to highlight and illustrate the specific properties of each identified taxonomical family, a few examples of visualization approaches pertaining to each of the four defined quadrants, while exhibiting a varying degree of intensity along the defined dimensions, will be provided in the following subsections. Furthermore, a short description and overview of each considered approach will also be presented alongside a brief summary of their respective strengths, weaknesses and limitations. Such task will be assisted by the further exploitation of the multidimensional model introduced in the first part of in section 3.1.6, which describes the relations linking the continua of visual consistency, precision, automatism, and interactivity. The degree of intensity featured by each approach concerning such dimensions, and the type trade-off exhibited by each visualization strategy, will therefore be introduced and briefly discussed. Finally, for each of the approaches that were selected to illustrate a specific taxonomical class, further examples concerning practical implementations and specific projects which have been relying on such visualization strategies will equally be mentioned.

81 80 Methodological approaches for the protection of cultural heritage in the digital age First quadrant: low virtuality and low interactivity The first taxonomical family which is considered for our categorization regroups the approaches that belong to the first quadrant of the two dimensional general conceptual model defined in section The visualization strategies that can be classified inside such portion of the conceptual space, generally exhibit a degree of virtuality and a degree of interactivity that are both ranging from insignificant to slightly below average. In other words, this first quadrant regroups the visualization attempts which are predominantly developed by employing traditional techniques and which are featuring relatively static capabilities for their final users. In order illustrate the typology of approaches that can be classified as such, a few examples, introduced by figure 7, are explored in the following subsections. Interactivity a 1. Restitution Drawings 2. Augmented Pictures 3. Scale Models Virtuality Figure 7: examples illustrating the first quadrant of the taxonomical space It is important to stress that the range of visualization strategies that are susceptible to belong to each introduced family is not exclusively limited to the cases that are explicitly mentioned hereunder, but can be extended with a theoretically infinite amount of further instances. In fact, each taxonomical family is intended as a general container for any approach that is exhibiting similar degrees of virtuality and interactivity. Therefore, the presented approaches are provided only as examples which illustrate the main characteristics exhibited by all the strategies that belong to such conceptual families Restitution drawings In the frame of an archeological dig it is generally standard procedure for the excavation team to define accurate drawings, to create precise illustrations, as well as to constitute a photographic documentation of the excavated items, as part of the recording process performed on site [LOCK03]. The content of these drawings can range from the creation of general maps, to the detailing of

82 Related work: taxonomy and review 81 architectural and sculptural elements [IZENOUR92], to the definition of plans and sections of the considered items [DEBERNARDI70]. In the specific case represented by pottery, for instance, since it is the archeological artifact most commonly found on excavation sites, the artifacts are fist cleaned and dried, then selected and recorded, eventually restored or glued, and finally, as last step, they are brought for drawing and inking [YILMAZ02]. Architectural drawing and sketching, as a way to quickly document the findings and visually record their state, features mainly two advantages other than their relatively short development cycle. First they offer the possibility to clearly identify small details, tiny decorations, or barely readable inscriptions recorded on site, which allows for an efficient classification of the excavated material. Secondly, it gives the archaeologists the chance to schematically compare each element in its entirety while offering a support for the definition of a personal mental reconstruction of their past appearance. Moreover, if skillfully executed, a drawing, or a sketch, can equally be employed as a medium to propose a visual archeological restitution of a specific heritage item in its time dependant context, i.e. exhibiting the visual appearance it was supposed to display at a certain time in the past, thus concretizing the results of the imaginative reconstruction performed by the archaeologist. As a few examples of hand drawn restitution drawings targeting highly complex buildings, we can mention, for instance, the lithographic work executed by the Fossati brothers on the Hagia Sophia edifice in Istanbul [HOFFMANN99], or the views of the scene of the Roman theatre of Aspendos, in the region of Antalya [LANCKORONSKI92]. Visual Consistency and Precision Interactivity Automatism Figure 8: restitution drawings, classification of the approach in white, modified position accounting for the possible application of CAD techniques, in gray However, such method heavily relies on the personal skills of the illustrator executing either the documentation or the restitution of the considered item. Therefore, the capacity to produce visualizations that systematically display a satisfactory visual consistency and a high degree of precision is generally limited. Moreover, this capability is further restricted by the fact that a handmade drawing, by its nature, cannot generally rival in photorealism with its real counterpart, and by the fact that it cannot integrate the full range of the geometrical detail that is featured by the considered real item. Furthermore, such visualizations do present a small degree of interactivity, since they are

83 82 Methodological approaches for the protection of cultural heritage in the digital age basically static images drawn on paper, and offer an unpractical support to implement further modifications to the chosen visualization. In other words, if it were necessary to display a visualized item from a different point of view than the one originally prepared, or if it were required to modify the drawing to implement and test some particular variations to the base concept, it would be necessary to create a new image from scratch. In order to alleviate such drawbacks, it is possible to support such tasks with computer assisted drawing and shape recognition tools to digitize and modify the drawn images: however, such possibility would impact on the localization of this approach in figure 7, where has been noted as point 1a, and in figure 8, where has been noted as a gray box, since it would give such strategy a slight increase in its degree of interactivity, due to the improved ease in modifying the results, and in its degree of virtuality, which is offered by the introduction of CAD techniques Augmented pictures With the term Augmented pictures, we refer to the creation of accurately executed partial restitution drawing, generally colored, which have been perspective matched and superimposed to a photography of the remains of a given item or ancient building. Such superimposition is generally achieved either by means of a transparent sheet of plastic upon which the drawing is painted, or printed, that can be flipped by the user or by means of a more sophisticated system in which two images, one with the augmentation and the other without, are printed on the opposing faces of a milled surface that allows only one image at a time to be seen by the user depending on its point of view. These images are generally employed for dissemination purposes in the frame of either short publications, didactic material or as postcards intended for tourists at archeological sites. As examples belonging to his category we can cite, for instance, the augmented reconstructions of Pompeii and Ercolanum prepared by Cozza and Staccioli [COZZA55], or the visualization attempts developed by Vision Roma, which have been included in several books [VISION ROMA95] and have been published as postcards. Visual Consistency and Precision Interactivity Automatism Figure 9: augmented pictures, classification of the approach This kind approach allows for the visual integration and the direct confrontation of selected restitution hypothesis with the present reality. As such, while sacrificing to a certain degree the maximum possible precision achievable, as it

84 Related work: taxonomy and review 83 is the case with restitution drawings, augmented pictures generally offer the users a slightly increased perceived visual consistency due the association of real and painted elements: this increase is achieved at the expense of longer development cycles and a reduced flexibility for the inclusion of further modifications to the final visualization due to the amount of detail and the specific techniques that are used by such approach. Furthermore, the possibility to offer the users the possibility to experience a cultural item in both its current state and in its restituted one, either on site or not, is accounted by a degree of interactivity slightly more pronounced than manual archeological restitution drawings Scale models With the term scale model, we refer to a time dependant physical reproduction, made out of wood, plaster or other construction materials, which visually represents a heritage item in either its current state, i.e. an exact replica of the real one, or in its restituted state, i.e. illustrating the appearance it would have exhibited at a specific time in the past, and which is built at a lesser scale than 1:1, i.e. life-size. The type of heritage item that can be the subject of a scale reproduction is rather broad, ranging from small artifacts, to siege machines and buildings, to complete models of entire cities. Such method of visualization is suited for dissemination purposes and as a tool for analyzing and studying the considered items while allowing the testing of specific physical and visual hypothesis concerning their supposed appearance or functioning. Since the eighteenth century, architects started to provide such models to scholars and students in order support their efforts aimed at better understanding the ancient remains which were used as reference for their construction. One of the first examples of the exploitation of such approach can be traced back to a famous cork-modeler named Antonio Chichi, who lived between 1743 and 1816, and whose craftsmanship allowed the production of a set of replicas representing several of the main sites of the ancient Rome [FRISHER04]. More important, these replicas were not exclusively intended as souvenirs for tourists, but were created to serve also as study aids similarly to the plaster casts that were made using famous Greek and Roman statues as base reference for the modeling [WILTON96]. Several other examples of scale models serving the same purposes, but proposing the visualization of heritages items in their restituted state, can also be mentioned. For instance, two of the most impressive and larger plaster models, offering the visualization of a restituted site, are available for public display at the Museum of the Roman Civilization located in Rome. These models, which respectively present a visualization of the ancient city of Rome restituted as it was during the early Republican period, and as it was during the era of Constantine I, were built at scale of 1:250 by Italo Gismondi, and required almost thirty years of work to complete [LIBERATI03]. As further examples belonging to his category we can also cite, for instance, Giuseppe Fiorelli s, model of the excavation site of Pompeii, built at the scale of 1:100, which is exposed at the National Archaeological Museum in Naples, Paul Bigot s model of the imperial Rome, which is located at the University of Caen, or the scale model of the ancient basilica of St. Peter s displayed in the Vatican museums.

85 84 Methodological approaches for the protection of cultural heritage in the digital age Visual Consistency and Precision Interactivity Automatism Figure 10: scale models, classification of the approach The approaches aimed at the visualization of a heritage property through the physical reproduction of a scale model, generally exhibit a mild visual consistency and precision, depending on the amount of detail enabled the by a chosen scale, the closer being to 1:1, the more detail can be included in the model, and by the type of surface material employed. Moreover, since very detailed scale models are usually large and heavy, they are usually only available at fixed locations for display, thus limiting the possibility to quickly move and reassemble them at a different location. Furthermore, while the development cycle for the creation and finalization of small scaled models, can be reasonable, the creation of larger physical visualizations, as it is illustrated by Italo Gismondi s plaster models of the ancient city of Rome, usually requires a conspicuous amount of time and effort to achieve completion. Additionally, representing heritage items by employing this kind of approach allows for an increased interactivity in respect to the previously mentioned visualization strategies, since it offers the users the chance to freely explore, study and experience the model by modifying their relative point of view in a three dimensional physical space. Finally, scale models equally offer scholars the possibility to visually test in a three dimensional space specific archeological hypotheses. However, while they require a lesser expense in terms of resources and time than the creation of a 1:1 replica, scale models do not provide the same physical precision in terms of their capacity to reproduce the same physical conditions, constrains and properties, that is exhibited by their real counterpart Second quadrant: low virtuality and high interactivity The second conceptual family that is employed by our taxonomy regroups the visualization strategies which belong to the second quadrant of the two dimensional general conceptual model defined in section The approaches that can be categorized inside this area of the conceptual space feature a low degree of virtuality while exhibiting a degree of interactivity which is ranging from slightly above average to pronounced.

86 Related work: taxonomy and review 85 2 Interactivity Physical Reconstructions 2. Interactive Scale Models 3. Live Experiments Virtuality Figure 11: examples illustrating the second quadrant of the taxonomical space This second quadrant generally regroups the visualization projects that are developed by employing traditional techniques and that are featuring relatively interactive capabilities for their final users. In order illustrate the typology of approaches that can be classified inside this family, a few examples, which are introduced in figure 11, are explored in the following subsections Physical Reconstructions By using the term physical reconstruction we refer to the visual representations of heritage items that feature reconstructive interventions, either minimal or extensive, which are applied on site while employing, wherever possible, original architectural elements and materials. Such approach is also referred as anastylosis and represents the attempts at visualizing a given heritage property in its restituted state by means of the physical integration between still existing remains present at the site and newly built, or reconstructed, structural elements. A striking example of such visualization strategy is represented by the interventions executed by Sir Arthur Evans at the archeological site of Knossos, where he led systematic excavations since 1900 [EVANS21]. Among the intervention that were directed by Evans we can mention, for instance, the controversial reconstructions that were applied to numerous frescoes, which were originally discovered as incomplete small scattered fragments, and that were subsequently reassembled and restored. Evans also directed the physical reconstruction and restoration of larger items, such as complete rooms within the Knossos palace while employing construction materials foreign to the original ancient Minoan architecture, such as steel-reinforced concrete applied on top of the ancient remains. On the basis of these examples, it is reasonable to expect that such methods have been extensively criticized due to the permanent nature and the over-interpretation that characterized most of these interventions, and by the fact that the proposed restitutions did not integrate a

87 86 Methodological approaches for the protection of cultural heritage in the digital age scientifically accurate or a historically pertinent vision of the past appearance of the ancient palace [ATHENA03]. This example illustrates an uncommon extreme case of invasive reconstructive intervention; however, such experiences can be executed in a more subtle fashion and could be targeted more at strengthening the preservation of the remains, though conservative structural interventions, rather than being aimed at the full restitution of the past appearance of the site. As examples of such alternative we can mention the interventions that were carried out at the theatre of Aspendos, located near Antalya in Turkey, the partial restitutions conducted on the Celsus Library in Ephesos, or the reconstructive and conservative interventions that can be applied to pottery fragments found during archeological excavations. In the particular case of Aspendos, for instance, the scene of the theatre and all the architectural elements that were presenting clear degradations or missing parts, but whose structural integrity was endangered, were not touched. Inversely, certain structures, such as some sections of the arcades located on top of the cavea of the theatre, which are sustaining the roof, were reconstructed using modern materials both to increase their structural stability and to clearly indicate to the visitor that such elements were not part of the original building and that were recently added. Physical reconstructions generally provide the visitors of an archeological site with the possibility to visually experience a highly consistent representation of the past appearance of heritage items. However, the historical accuracy, and global precision, of such strategy can sometimes be questionable if not supported by enough evidence. Moreover, while it could be acceptable, to a certain extent, to intervene on selected items still featuring enough data to allow the reconstruction process to be executed by employing a process of analogy, it is otherwise unsafe to implement such reconstructive interventions directly on top of an archeological excavation site, since such approach would both preclude the access to further findings and obscure the genuine remains. Visual Consistency and Precision Interactivity Automatism Figure 12: physical reconstructions, classification of the approach Regarding this last element, it is important to note, that the execution of a questionable reconstruction on a poorly documented site would probably prove to be more harmful than helpful for a visitor, since it could convey the feeling that such restitution has been done following clear and precise data. Furthermore,

88 Related work: taxonomy and review 87 such approach could equally generate some confusion concerning the necessary distinction that should be made between the reality of the remains and the hypotheses that have been formulated regarding the past appearance of the cultural property. Consequently, the presentation of a physical restitution of such an archeological site, could crystallize in the mind of the observer the idea that the presented item, in its restituted state, represents the absolute reality of the past, therefore precluding the possibility to consider other alternatives about the supposed appearance of the site. Additionally, beside offering a high degree of visual accuracy, especially if during the restitution process similar materials than the ones that were originally used in the past were employed to restitute the appearance of the considered items, such visualization attempts generally feature a mild degree of interactivity, since they allow the users to fully experience, visit and explore a cultural item at a real life scale of 1:1. Finally, as it is intuitively obvious, this type of visualization strategy is usually accompanied by a very low degree of automatism, since the physical construction of a detailed life-like restitution, based on specific remains, as it is the case for the Frauenkirche in Dresden, generally requires a great amount of time and effort to complete Interactive scale models The visualization attempts use this type of strategy share comparable properties with the ones that rely on static scale models, which were overviewed in section Interactive scale models are, in fact time dependant physical reproductions, made out of wood, plaster or other construction materials, representing a heritage property in either its current state or in its restituted one, which are generally built, for practical reasons, at a lesser scale than 1:1. Additionally, such scale models also present specific moving parts that can be manipulated or activated by the users: these mechanisms mimic and reproduce, at a lesser scale, the functioning that is inherent to some specific heritage items, thus visually illustrating their role, purpose or construction method. This type of scale model is generally employed for the visualization of a single entity, such as an ancient water pit apparatus, a particular building, or a small complex, and is usually focused on a specifically chosen mechanism or function, such as the heating system of the Termae in Rome, the mechanisms allowing the aqueducts to convey water in Roman houses or the raising and lowering system of the velum in Roman theatres. Moreover, such visualizations are usually presented as cut away views of the considered objects, which reveal to the observers parts of the item that would have been otherwise concealed and which allow for a clearer illustration of the mechanisms targeted by the visualization. An example of such interactive scale models are the installations that can be often observed in scientific expositions such as the ones presented at la Cité des Sciences in Paris [CDS08]. However, this approach can be extended to also visualize cultural items; as such, we can mention, for instance, the interactive reproductions of the inventions made by Leonardo da Vinci that can be observed and interactively explored in the National Museum of Science and Technology [NMST08], located in the city of Milan. As a further example, we can also cite the interactive cut-away scale models that are exposed at the Museum of the

89 88 Methodological approaches for the protection of cultural heritage in the digital age Roman Civilization, located in the city of Rome, which illustrate the functioning of specific cleverly built architectural elements commonly employed by ancient Roman engineers. Visual Consistency and Precision Interactivity Automatism Figure 13: interactive scale models, classification of the approach This type visualization strategy, analogously to the one previously explored in section , usually features a mild visual consistency and precision, depending on the amount of detail included in the model, the materials employed and the scale chosen for the reproduction. However, since such models generally target smaller items, it is not uncommon to see 1:1 interactive reproductions of specific artifacts, such as the previously described Leonardo s machines. In such cases, a slight increase in the visual consistency exhibited by such visualization attempts can be considered. Additionally, since such scale models are generally intended for direct manipulation by the visitors, their capacity to establish a direct relation with the users their capacity to raise their interest and curiosity, and the possibilities they offer to the observers to manipulate them, and thus interactively understand their functioning, are accounted by an increased degree of interactivity. Finally, we can note that such increase in visual consistency and interactivity is balanced by a weakened degree of automatism. The implementation, conception and construction of the interactive elements that need to be included in the model is generally achieved at a greater expense, in terms of time and effort, than what it is usually required for the creation of a simple static scale model Live experiments Live experiments constitute a special case of interactive scale model where the replicas are always built at a scale of 1:1. Such models are not generally meant for public display at museums and usually do not feature the presence of devices acting as interface between the users and the item to enable their interactive qualities. The highly interactive aspect of this approach resides in the fact that live experiments are directly experienced by their users throughout all the phases that play a role in their creation, from the initial construction phase, to the complete manufacturing process, to the finalization and testing of the reproduced model in real life conditions. Additionally, since the creation of a live

90 Related work: taxonomy and review 89 experiment generally requires an active participation from its users, rigorous historical reenactments, such as for instance a live reconstruction of an ancient Greek battle [HANSON01] scientifically recreated in location with historically coherent costumes, can equally be mentioned as belonging to this kind of approach. Visual Consistency and Precision Interactivity Automatism Figure 14: live experiments, classification of the approach This type of visualization strategy, which is aimed at the replication of ancient artifacts or known past activities, is best suited, and mostly used, for hypothesis testing in the research field of experimental archeology [BECKER98]: the areas of interest of this field are centered on the analysis and on the direct experimentation, in controlled conditions, of both the behavior exhibited by humans in the past and the technologies that they were employing, including the manufacturing processes that were used to create specific artifacts. An experimental archeology test, or a live experiment, can thus be described [COLES73] as an attempt to reproduce specific tools, artifacts, object or edifices, as well as the processes that were responsible for their degradation or destruction, through a direct experimentation conducted under a set of conditions that match as close as possible the ones that were present at the place and time to which the targeted item originally belonged to. Field experiments represent a systematic approach which can be employed to test and evaluate the methods, the techniques, the assumptions, the hypotheses and the theories that have been formulated at all the different levels of an archaeological research or study [INGERSOLL77]. As a few examples that implement this visualization approach, we can cite, for instance, the creation and the further testing during a sea trial of a model Viking boat that was built by the Viking Ship Museum of Roskilde by using ancient manufacturing methods in the frame of the Thoroughbred of the Sea project [SORENSEN06], as well as the post use-wear analysis replication experiments, based on the works of [HEALEY92], that were performed by the Institute of Archaeology of the University of Oslo [IAKK08]. As a further example, we can mention the impressive work carried out at the site of Little Buster, where a permanent working ancient farm was constructed in order to experimentally investigate past farming practices [REYNOLDS99].

91 90 Methodological approaches for the protection of cultural heritage in the digital age While sharing, to a certain extent, similar characteristics with interactive scale models and physical reconstructions, live experiments, as a way to visualize, present, study, test, understand and experience first handedly the hypotheses regarding the appearance, functioning, or manufacturing process related to a specific heritage item, do exhibit specific advantages and drawbacks compared to such approaches. As such, we can note that live experiments exhibit a more pronounced degree of interactivity, in respect to simple interactive scale models, while they retain the high visual consistency that is featured by physical reconstructions. Moreover, although they generally require less resources, in terms of time and efforts, than on site restitutions built on existing remains, experimental archeology field tests however need longer development cycles that the construction of interactive scale models. Furthermore, it has to be mentioned that the relative global precision that can be achieved by such visualization strategy greatly depends on the reliability of the sources that are used to create the experimental models and on the scientific rigorousness that is applied to the testing process. As a final remark, we can also note that, whenever a reconstruction attempt produces invalid or dubious final results due a lack of solid information, the impact on the original heritage properties that are used as a base for the visualization is less dramatic than in the case represented by physical reconstruction, since the replicas are built from scratch without physically integrating the model with the actual archeological remains Third quadrant: high virtuality and low interactivity The third category of visualizations that is employed by our classification regroups the strategies that belong to the third quadrant of the two dimensional general conceptual model defined in section The approaches that can be categorized inside this specific area of the taxonomical space usually feature a pronounced degree of virtuality while exhibiting a degree of interactivity that ranges from irrelevant to slightly below average. 1. Renderings 2. Digital Catalogs 3. Digital panoramas Interactivity 2 3 Virtuality 1 Figure 15: examples illustrating the third quadrant of the taxonomical space

92 Related work: taxonomy and review 91 This second quadrant regroups the visualization strategies that predominantly employ digital technologies and which are exhibit relatively static capabilities for their final users. In order illustrate the typology of approaches that can be classified as belonging to such category, a few examples, which are introduced by figure 15, are presented in the following subsections Renderings created using Computer Graphics By employing the term rendering we refer to any computer generated image, or series of images, that depict a non augmented visual representation of a cultural heritage property in either its restituted or current state. Such visualizations are produced by employing a three dimensional model of the considered items, which have been manually modeled or digitally scanned, thus enabling a variable degree of geometrical precision in respect to their real counterpart. The appearance that the modeled items exhibit in the final scene is defined by employing surface material simulations, lighting computations and mathematical models that describe their physical attributes and behavior. Once a virtual scene is completely defined, populated by the modeled objects and eventually animated, a snapshot, or a series of snapshots of the virtual world can be computed from a chosen point of view and then saved as a digital image, or as a sequence, for later display. To achieve a visually consistent and historically precise 3D virtual restitution of a selected heritage property different aspects are critical [FONI02], such as, for instance, the choice of the appropriate modeling techniques, the preparation of the illumination models, the accuracy applied to the texture and material creation processes, and the reliability of the hypotheses upon which is based the chosen virtual restitution. As a few examples of such approach we can cite the radiosity processed movies presenting the virtually restored Church of SS. Sergios and Bacchos, located in Istanbul, made in the frame of the CAHRISMA project [PAPAGIANNAKIS01], or Paul Debevec s photorealistic visualization of the Parthenon in Athens generated by means of high dynamic range image based lighting techniques [DEBEVEC04]. Additionally, as further examples, we can mention the computer generated illustrations of the Trelleborg fortress made by Jacques Martel for the Journal of Medieval History [VIRTUHALL08] and the virtually animated reconstructions of the seven Wonders of the Ancient World featured in a documentary belonging to the collection Lost treasures of the Ancient World [KULTURFILMS00]. Computer generated images and movies, like photographs and filmed footage, are characterized by an almost irrelevant degree of interactivity, which is exclusively limited to the emotional response that such medium could eventually stimulate in the viewer. Such visualizations are, in fact, intended to be watched by the observers passively, without requiring any interactive relation between the user and the medium representing the cultural item. Additionally, the creation of computer generated movies, and even single detailed rendering, is usually a process that requires long development cycles, since a lot of time is spent for the creation of the virtual 3D models and later for the execution of the computations that are necessary to produce the final images. While such rendering time might at first appear reasonable when considered for a single image, whose computation generally implies a few CPU hours to achieve

93 92 Methodological approaches for the protection of cultural heritage in the digital age photorealistic qualities, it is evident that for a movie, in which at least twenty four images have to be computed for each second of animation, several days might be required to produce a final result. Visual Consistency and Precision Interactivity Automatism Figure 16: renderings, classification of the approach It has also to be noted that the rendering pipeline is usually a trial and error process in which several parameters have to be tested to determine the most suitable combination ensuring good visual consistency and reduced computation times. This amount of time can generally be traded-off for a decreased visual consistency in the final image or for a less detailed geometrical precision of the visualized items. A less detailed geometrical model would, in fact, require fewer resources for its modeling and rendering phases, and simpler physical simulation models and rendering algorithms would more quickly produce less visually consistent images. As an important side note, it is essential to mention that such visualization strategy can produce, given the appropriate conditions, expertise and tools, comprehensive and precise restitutions that can provide the viewers with additional elements not exclusively limited to the geometrical reproduction of the past appearance of the considered items. The addition of historically consistent virtual humans [THALMANN04] and the implementation of narrative dimensions [CAVAZZA03], would, for instance, allow for a better understanding of the past use, function or specific social role of the visualized cultural item, and the inclusion of specific ambiences, atmospheres or dramatic elements would permit the production of more precise cultural heritage simulations [FONI07]. Moreover, given the computing power delivered by modern processors, complex and accurate non interactive simulations could also be performed by employing three dimensional models of the virtually restituted cultural items along with their physical attributes. As such, the valuable option to execute virtual experimental archeology investigations, and virtual field trials, would then be offered [THALMANN07], thus allowing for a decrease of the development cycles that usually characterize such traditional approaches. Virtual archeology simulations [REILLY90] would not feature however a comparable degree of interactivity as live experiments, since the visualization of each of the computed results would necessarily require an amount of processing time ranging from a few minutes to a few hours.

94 Related work: taxonomy and review Digital catalogs With the term digital catalog we refer to the visualization projects, generally undertaken by museums, aimed at publishing for public consultation in electronic format the information that is relative to a specific collection of cultural items in their possession while integrating in an organized manner both the graphical illustrations, either photographic, rendered or hand drawn, and the textual documentations that is linked to each presented artifact. As a complementary side note, we would like to mention that the term virtual museum could be used to partially describe such approach, since such virtual spaces originally employed the World Wide Web and the HTML technologies to display simple hyper-linked text accompanied with graphics [JONES02]. However, despite the fact that it is not uncommon to still encounter such form of virtual museums, as exemplified, for instance, by the virtual museum of Japanese Arts [VMJA08], for the museum of Canada [VMC08] and for the English national archives [UKNA07], the rapid advances made in computer graphics and the democratization of high speed internet connections has nevertheless encouraged and finally allowed the integration of more complex data into such virtual spaces [TSICHRITZIS91]. This data can range from clips, animations and videos, to digital panoramas, interactive three dimensional visualizations [MITCHELL99] and structured databases [PAQUET01]. Moreover, it is possible to include, in order to expand the application range of such general term, other specific digitally based visualization techniques, such as haptic models for instance [MCLAUGHLIN00], which could be used as support for the presentation of heritage items in the frame of a museum exposition. Visual Consistency and Precision Interactivity Automatism Figure 17: digital catalogs, classification of the approach All these examples however represent more a specific implementation of particular visualization technologies and digital techniques, which are applied to extend and enhance the real space of a museum, than a unified concept that can be presented as a single approach to visualize cultural heritage properties. Since such strategies exhibit very different characteristics, qualities and attributes, they would therefore occupy different spots inside our taxonomical spaces. For this reason, as an example that illustrates the third quadrant of our conceptual space, we prefer to refer exclusively to the limited case represented

95 94 Methodological approaches for the protection of cultural heritage in the digital age by digital catalogs, which corresponds to a subset of all the possible instances issued from the more generic concept of virtual museum [LEPOURAS01]. The visualization of heritage items though the creation of digital catalogs is characterized the fact that the real objects themselves are accessible to the users only as digital proxies. Such digital representations do not display a high degree of virtuality since often they are not completely modeled, i.e. the computer detains only limited and partial information regarding the content and meaning of such images; thus it is important to mention that if a catalog exclusively exhibits the presence of digital photographs without any renderings nor additional information or attribute accompanying the images, its classification could almost slide inside the first quadrant of the taxonomical model. Moreover, although digital catalogs exhibit a fairly good degree of visual consistency and a sufficient degree of precision, since they are often based on photographic documentations, and while they provide simple methods to browse and search the vitalized cultural items according to specific attributes that describe their characteristics, they do not, however, usually offer a remarkable degree of interactivity. Finally, since the primary goal of such approach is frequently limited to provide a documentation of the current appearance of the considered items, although computed restitution rendering can occasionally be featured alongside the illustrations of the artifacts, the creation and finalization of a digital catalog that presents an average amount of items, accompanied by their corresponding images and textual information, is often a process that requires a low degree of automatism Digital panoramas The term digital panoramas describes the approaches that employ a set of digitally taken neighboring and partially overlapping snapshots, which have been processed, seamlessly stitched together [AUTODESK08] and mapped to the surface of a sphere, a cube or a cylinder, to allow the viewer to experience a chosen space or object, either virtual, real or augmented [VERGAUWEN04], by rotating the mapped shape, therefore giving the user the illusion that he is standing inside of such space or in front of the visualized object. Special software, such as for instance the QuickTime VR technology [SHENCHANG95] developed by Apple [APPLE08], or the IPIX system [IPIX08], are employed to generate and display the mapped panoramas, as well as to allow the manipulation of its orientation and zoom level. Furthermore, it is also possible to include and define in such images specific areas, called hotspots, that can contain hyperlinks to other digital panoramas, web pages, documents or annotations [ZARA04], thus defining a virtual space organized in a node structure that can be explored by the viewer. This approach was adopted to create, for instance, the virtual tour of the Cathedral of Milan and the Basilica of St. Peter in Vatican [PAIANO01], the visualization of the Santa Maria degli Angioli Church located in Lugano [VRWAY08], as well as for the constitution of the virtual environment of the Edinburgh s Royal Mile [WRIGHT99]. Additionally, as further examples, we can also cite the augmented visualization of the restituted Byblos theatre presented in [ELKHOURY06], the virtual restorations of the Al-Suq al-faransawy building,

96 Related work: taxonomy and review 95 located in Alexandria [ADEL05] and the augmented virtual reconstructions of the landscape of the Ename site [VERGAUWEN05]. Visual Consistency and Precision Interactivity Automatism Figure 18: digital panoramas, classification of the approach As introduced in section , this particular visualization strategy can be used to create digital representations of heritage items that are suited for a virtual museum project. In fact, digital panoramas have been often employed and widely used for such projects, as it is the case, for instance, for the museum of the Oriental Institute of the University of Chicago [OIUC08] since their relatively short and inexpensive development cycle can offer comparable degrees in both visual consistency and precision as digital catalogs while allowing at the same time an increased interactivity. However, despite the presence of hotspots, the global level of interactivity and virtuality offered by digital panoramas is nevertheless relatively limited, since the users are not allowed to freely explore and manipulate the visualized spaces and objects, and are confined to the exploration of a series of nodes containing partially modeled static images that can only be rotated and zoomed. As a concluding remark, we would like to mention that the visualization strategies that have been presented to illustrate this particular quadrant of the taxonomical space are, more often than not, employed together in the frame of a virtual museum experience. As it is the case, for instance, for the didactical CDROM Gli Etruschi [RNM00]: as such, digital panoramas, renderings, computer generated movies, and digital catalogs can be regrouped on the same medium to combine their respective strengths and offer their users an enhanced and more varied visual experience Forth quadrant: high virtuality and high interactivity The last category of approaches that is employed in the frame of our categorization regroups the visualization strategies that belong to the fourth quadrant of the two dimensional general conceptual model defined in section The strategies that can be classified as belonging to this specific area of the conceptual space generally exhibit degrees of virtuality and interactivity which are ranging from slightly above average to strong. This second quadrant regroups the visualization attempts that predominantly employ digital

97 96 Methodological approaches for the protection of cultural heritage in the digital age technologies and which are exhibiting relatively high interactive capabilities for their final users. In order illustrate the typology of strategies that can be classified as belonging to this taxonomical family, a few examples, which are introduced in figure 19, are presented in the following subsections Interactivity 1. Real time VR simulations 2. Stereoscopic visualizations 3. Videogames Virtuality Figure 19: examples illustrating the fourth quadrant of the taxonomical space Real time virtual reality simulations The visualization projects that rely on real time virtual reality simulations share a few characteristics with computer generated renderings, which were previously presented in section Consequently, with such term, we refer to any computer generated visualization that depicts a real time non augmented visual representation of a completely modeled cultural heritage item, in either its restituted or current state, which was produced by employing a three dimensional digital description of the geometrical attributes of the considered item. However, inversely to what happens for the production of static renderings, where a fixed or moving point of view is first defined in the virtual world and all the images corresponding to such camera positions are computed and saved as digital pictures or movies, the visual results of real time virtual reality simulations have to be computed and displayed on the fly. While the finalization of a detailed static rendering usually requires at least several minutes of computing for each produced image, real time visualizations need to generate and display a minimum of twenty five to thirty rendered images per second. The level of interactivity that such applications exhibit can therefore be remarkable, since real time applications provide their users with the possibility to freely manipulate the visualized items and to directly explore the modeled virtual worlds at will, while generating a real time visual feedback for each of their performed actions and choices. Among the examples that implement such approach to interactively visualize heritage items we can cite, for instance, the real time virtual reconstruction of Notre Dame in Paris [DELEON00], the immersive virtual tour of the Sienna cathedral [KNOPFLE00], the real time animation of King Ashur-nasir-pal II

98 Related work: taxonomy and review 97 [KIM01], the visualization of the Temple Site at Phimai [LEVY01] and the online exploration of the Louvre virtual museum available at [LOUVRE08]. Visual Consistency and Precision Interactivity Automatism Figure 20: real time VR simulations, classification of the approach While real time interactive virtual reality simulations can exhibit fairly photorealistic qualities, as demonstrated by the inhabited real time virtual restitution of the Aspendos Theater presented in [THALMANN07], their overall geometrical precision and visual consistency is nonetheless generally limited by the necessity to provide a real time feedback to the user. Therefore, real time visualizations normally have to rely on specific solutions, such as bump mapping [BLINN78], normal mapping [COHEN98], pre-computed radiance transfer [PAPAGIANNAKIS05] and light map based strategies [ZHUKOV98] to compensate for the impossibility to interactively display complex and very detailed models or to visually approximate the results that might otherwise be obtained by precise and accurate non interactive physical simulations, such as radiosity based light computations. However, in order to enhance by such means the visual consistency and the perceived precision that is exhibited by a real time simulation whilst retaining its interactive qualities, as well as to allow the implementation of a narrative dimension accompanied by the inclusion of specific ambiences, atmospheres and dramatic elements, it usually necessary to sacrifice a certain degree of automatism. Conversely, whether necessary, it is possible to shorten the development cycle that characterizes such visualization attempts, whilst maintaining a high degree of interactivity, by trading-off some precision and visual accuracy Stereoscopic visualizations The term stereoscopic visualization refers to any technique which is capable to stimulate in the viewer the illusion of real 3D depth by employing two dimensional data: such result is typically achieved by separately presenting to each eye of the observer a slightly different image which is then reconstituted in his brain, thus generating a perceived depth in the visualized scene. One of the fist examples of machine that employed such technique is the stereograph built by Sir Charles Wheatstone in 1838, whose functioning was based on a system of prisms and mirrors to convey to each of the eyes of the user the intended

99 98 Methodological approaches for the protection of cultural heritage in the digital age image. A popular implementation of such concept are anaglyph images, where the viewed subject, generally composed by two superimposed color layers which have been off-set, is observed with paper made glasses featuring a red and a blue filter [SHOUKANG98]. More recent techniques that exploit such basic concept [BOURKE00] include front and rear projection systems which are employed to convey the correct image to the intended eye by using an active or a passive stereo strategy while respectively featuring the implementation of shutter glasses and polarized glasses. Additionally, research and manufacturing efforts have also been directed toward the development of auto-stereoscopic displays [DODGSON99] [HALLE97] [JONES07], which allow the viewer to experience a full three dimensional visualization that directly conveys the perception of depth without the need of specific glasses. Such systems however can prove rather expensive, depending on the technical solution that is chosen to support the projection. The available options range from the adoption of an immersive CAVE or ImmersaDesk systems [EVL08], to the installation of a single high resolution and high frequency projector equipped with a polarizing filter capable of displaying 120 images per second [CHRISTIEDIGITAL08], to more reasonably priced setups, such as for instance the coupling of two linked polarizing projectors that have been offset [CLIBURN05], the use of high frequency LED television screens coupled with first party propriety 3D glasses, or the in house creation of low cost solutions [ZELLE04]. Visual Consistency and Precision Interactivity Automatism Figure 21: stereoscopic visualizations, classification of the approach These stereographic techniques can be used to visualize several different types of source material, such as renderings, computer generated movies, photographs or digital panoramas. The visualization of Angkor, located in Cambodia, presented in [KENDERDINE04], for instance, employs a cylindrical Virtual Room [VROOM08], composed of eight rear projected screens, where a passive stereoscopic panorama is employed to visualize the heritage site to the viewers. However, despite the broad spectrum of possible implementations that are allowed by such strategies, in this section we are only considering a narrower application range for the technologies and methods related to the production of stereographic visualizations. In this respect, while considering the approaches that exemplify the fourth quadrant of our taxonomical space, our

100 Related work: taxonomy and review 99 attention will be focused exclusively on the applications and the techniques that enhance real time virtual reality simulations. Stereoscopic approaches can therefore be considered as an extension of the strategies that were presented in section As a few examples of interactive real time applications that implement such techniques we can cite, for instance, the virtual reconstitution of the ancient city of Seorabol site in Korea [PARK03], the artistic heritage reproductions presented in [LIVATINO06], as well as the simulations developed for the foundation of Hellenic world introduced in [GAITATSES01]. Additionally, we can also mention the usability study performed by [ROUSSOU99] targeting the implementation of stereographic devices in the frame of public museum expositions and the inhabited real time virtual restitution of the Hagia Sophia edifice as presented in [FONI07]. All these applications feature very similar properties, advantages and limitations as the ones introduced in section ; however, a few minor differences should be mentioned. The level of virtuality exhibited by such visualizations, as well as their overall perceived visual consistency can be considered, for instance, as slightly superior to plain real time virtual simulations. The capacity to stimulate in the viewer the perception of real depth requires the availability of an greater amount of digital knowledge concerning the displayed items to allow the system to correctly generate the stereographic pairs and, at the same time, it enhances the photorealistic qualities exhibited by the visualization: the ability to trigger the suspension of the user s disbelief is therefore strengthened. To produce a stereographic projection that can be visualized at a stable interactive frame rate it is however necessary to double the number of computed images, since each second of simulation requires the separate rendering and display of at least twenty five frames for each eye. To allow such level of computing performance it is generally required to decrease the geometrical complexity and precision exhibited by the visualized scene, since the rendering of a simpler model would necessitate less computing resources and therefore would be completed more quickly. Furthermore, the level of interactivity inherent to the applications that implement such approach can also be considered as slightly inferior to the level that is featured by simple real time simulations. While plain real time simulations can generally be individually accessed, displayed and freely used by their intended audiences on a personal computer, thus allowing a direct interactive relation between the application and its users, stereographic systems, which usually are either cumbersome or expensive, are inversely indirectly experienced, more often than not, by groups of people in specifically equipped environments, such as a CAVE system, a movie theatre, or a small projection room. In such cases, the overall visual experience and the navigation of the scene are generally driven by an external component, such as a scripted sequence of events or an operator which is manipulating the point of view that is presented to the observers. The overall level of interactivity that is subjectively perceived by each user is therefore weakened, since the feeling of being directly involved in the visualization, the sensation of being able to interactively modify the presented elements and the impression to enjoy a unique and personalized experience are diminished.

101 100 Methodological approaches for the protection of cultural heritage in the digital age Videogames Videogames have become in the last decade a widely diffused and popular medium for mass entertainment. Historically, such software applications have often been, and more as ever today, on the technological cutting edge of what is possible to achieve with the available digital technologies [SQUIRE03] [SQUIRE03b]. Videogames may be classified in different genres [GAL02] that exhibit specific design choices, specific technological implementations, which may range from digital panoramas to real time interactive simulations, and specific game play mechanics. Furthermore, it is not uncommon for such games to feature as esthetical backdrop, as constitutive contextual element or as supporting component for their narrative dimension a given historical period. As a few games, among the many available, that can exemplify such practice we can cite the city building video game Imperium Civitas [HAEMIMONTGAMES08], which is set in the ancient Rome, the infiltration game series Assassin s Creed [UBISOFT08], which in its first installment uses as historical context the period of the firsts Crusades, and the point and click adventure game Ankh [DECK13 08], which places the player in the ancient Egypt. Additionally, we can also mention the strategy game series Medieval Total War [CREATIVEASSEMBLY08], which stages playable setups of famous historical battles in the Europe of the XIth century, and the turn based strategy and simulation game Civilization [FIRAXIS08], which retraces the development of a whole civilization from the iron age to the space conquests while providing information on specific inventions and famous heritage edifices that the player can build to obtain specific in game advantages. However, since the primary objective of such medium is generally to entertain and engage the player and not to inform, teach, educate or disseminate information relative to heritage items, the cultural value that is featured by historically contextualized videogames is often a byproduct of their narrative choices and implemented scenarios. Thus, the cultural heritage virtual models that are included in a commercial videogame are usually lacking the sufficient degree of scientific rigor and precision to allow such visualizations, despite their generally good visual consistency, to be considered as a reliable way to accurately represent a heritage property. Moreover, the creation of a fully featured videogame that implements real time simulation technologies matching today s high visual standards requires the mobilization of large teams, the investment of far from negligible capitals and a considerable amount of time and effort to reach completion after a development cycle usually spanning across several months. As a side note, we would like to mention that videogames, as cultural heritage visualization tools, can be considered as exhibiting a slightly inferior degree of virtuality and interactivity than other previously presented approaches. A game experience, in most of the cases, is in fact structured around a set of narrative elements whose role is to provide a pre-determined set-up that offers a sense of meaning to the actions performed by the user [KLEVJER01]. The player is therefore frequently confined to the pursuit of the game objectives and to the witnessing of specific events while remaining in the frame that is defined by the narrative boundaries and by the spatial freedom that is offered in the context of the modeled playing environment. In this respect, the liberty of the user to

102 Related work: taxonomy and review 101 interact as he sees fit with the visualized environments and with the virtual representation of the objects that have been included in the game is limited both by the design choices and the specific narrative structure that have been implemented to drive the game experience. In other words, a player is not usually allowed to manipulate or to establish a direct interaction with all the items and the elements present in a videogame, since most of them are placed in the simulation just for decorative purposes as they are not fulfilling any narrative or game play objective. Similarly, during a game experience, a user is generally not allowed to explore areas and visualize items that are not supposed to be interacted with in the frame of the game structure. In an adventure game, for instance, where the backgrounds are static partial three dimensional models or static renderings that feature interactive hotspots, the user won t have the possibility to enter any house or manipulate any object that is visualized in a given area of the game world. Conversely, the player will be allowed to interact only with some predetermined elements by choosing an appropriate action among a proposed set of options, thus limiting the overall level of interactivity featured by the simulation. Visual Consistency and Precision Interactivity Automatism Figure 22: videogames, classification of the approach Although the narrative strategies and game mechanics that are employed in highly structured game experiences can reduce to a certain degree the overall freedom that is offered to the players, videogames nevertheless possess the potential to compensate for such shortcoming by providing their users with a strengthened perception of subjectively meaningful interactivity. Such compensation is not based on the absolute degree of freedom that is offered by the simulation, but is rather rooted in the possibility given to the players to perform and experience meaningful contextual interactions in the frame of the partially modeled game world. Considering the large success of videogames, and their capacity to engage their users [DICKEY05], since they offer a powerful mean to captivate, engage and channel the their attention, it is conceivable that the application of specific tools and technologies that were originally developed for entertainment purposes could benefit cultural heritage visualizations. The exploration of the educational potential of videogames [DEAGUILERA03], as well as the opportunity to employ them as tools to engage and stimulate learners while introducing specific

103 102 Methodological approaches for the protection of cultural heritage in the digital age historically related topics, has been documented, for instance, in [SQUIRE04], where a field experience has been conducted employing the commercial game Civilization [FIRAXIS08] as a didactic support for world history learning. Additionally, such approach was also implemented in the Virtual Reality Notre Dame Project [DELEON00], which featured the implementation of the first iteration of Epic s Unreal commercial engine [UNREAL08] to drive the real time simulation of the presented spaces. However, a more profitable approach would consist in actually studying the core mechanics as well as the interaction strategies that were successfully implemented in videogames: extending and adapting their application to virtual heritage environments [CHAMPION04] would benefit virtual heritage visualizations projects. The adaptation of such methods to visualize cultural properties would offer the valuable possibility to enhance the overall engagement experienced by the public and would provide an efficient medium to inform, teach, educate, and disseminate the information relative to our cultural heritage. Amid the heritage simulations that integrate such game mechanics we can cite, for instance, the multi-player role-playing game Revolution, which depicts the historical events that took place in Colonial Williamsburg, located in Virginia, during the American Revolution of 1775 [EDUCATIONARCADE08]. Additionally, we can also mention the location-based pervasive learning adventure game Anna's Geheimnis [SCHMIDT06], in which the user physically explores the Ilmpark, located in Weimar while employing a GPS capable Palm to collect clues, manipulate virtual objects and access information related to the cultural background of the site Mixed approaches In this section a few selected approaches that occupy special locations of the two dimensional taxonomical space defined in section will be presented. 1 3 Interactivity 3 Virtuality 2 1. Real time AR simulations 2. Digitally augmented movies 3. Semantically supplemented 2D photographs 3a. Semantically supplemented 3D models Figure 23: examples illustrating a few special cases in the taxonomical space

104 Related work: taxonomy and review 103 The strategies that belong to these particular areas of the conceptual model exhibit a mixed degree of virtuality [MILGRAM94] coupled with a degree of interactivity that ranges from insignificant to very high. This specific family of approaches regroups the visualization projects that combine and mix both digital and traditional technologies, and which feature varying interaction capabilities in respect to their final users. In order illustrate the typology of strategies that can be classified as such, a few short examples, which are introduced in figure 23, are presented in the following subsections Real time augmented reality simulations One of the goals that can be pursued by a heritage visualization that implements virtual reality related technologies is to provide their users with the feeling they are present at significant places and times in the past while stimulating a variety of senses to enhance their perception of what it would have felt like to be there. To achieve such end, a possible approach is constituted by the use of augmented virtual reality real time simulations, or augmented reality systems, which combine the concept of augmented pictures, previously introduced in section , with the techniques that are used for virtual reality real time simulations, which were overviewed in section Augmented reality systems, which are extensively detailed and described in [AZUMA97] [AZUMA01] [OHTA99], enhance the real world with computer generated virtual objects that appear to seamlessly coexist in the same space as the real ones. Such systems allow the users to visualize in real time a real environment on top of which virtual objects are interactively overlaid and integrated with the real ones that are present in the observed scene. In order to properly align the rendered virtual elements to the ones that are real it is necessary to determine the exact point of view from which the user is observing the objects and locations that should be virtually augmented. Furthermore, it is also essential to establish a correspondence between the real space and the virtual one to allow the positions of the virtual elements to properly match with the real environment. Visual Consistency and Precision Interactivity Automatism Figure 24: real time augmented reality simulations, classification of the approach To fulfill such requirements, camera tracking and match moving techniques are usually employed [DOBBERT05]. As such, the real environment is generally

105 104 Methodological approaches for the protection of cultural heritage in the digital age captured as seen by the user, by means of a small portable device such as a webcam for instance, while the recorded live images are directly fed to a computer whose role is to analyze in real time such images, either by employing a marker based shape recognition approach [LIAROKAPIS02] or a markerless image based strategy [LOURKAKIS03], and define a virtual camera that matches the position, the movement and the orientation of the real one. The registered virtual camera is then employed to align a three dimensional virtual scene that contains the virtual augmentations to the real environment, and it is used to define the point of view from which the virtual space will be visualized. Finally, the rendered elements are superimposed in real time to the live images and presented to the user by means of head mounted displays or stereoscopic devices. Since the process of camera calibration and motion tracking is usually computationally expensive, the simulations that employ such techniques generally exhibit simplified geometrical models to achieve real time capabilities, thus sacrificing to a certain degree their overall visual consistency and precision. For the same reasons, though some attempts have been made to approximate in a real time augmented environment complex illumination models that mimic the lighting featured by real scenes [PAPAGIANNAKIS05], such simulations often implement simple lighting algorithms, thus involving a weakened visual consistency. Furthermore, since the camera motion tracking process relies either on markers that are positioned directly in the real scene or on pre-processed image based data to determine the correct spatial position of the rendered elements, the augmented reality experiences which are offered to the user are generally confined to the exploration of specific limited areas. Additionally, such visualizations rarely provide their users with the capacity to directly manipulate and contextually interact with the virtually augmented elements, thus offering an overall slightly diminished degree of interactivity as what could be offered by virtual reality simulations. Despite such shortcomings, augmented reality restitutions can offer interesting possibilities and useful applications due to their fairly good interactive capabilities and their capacity to visually stimulate the viewer s engagement. Their uses can range from cultural heritage on-site visualization and exploration tools, to visual supports to enhance museum experiences [WOJCIECHOWSKI04] [HALL01], as wells as to serve as an edutainment medium [LINDT03]. As a few cases that exemplify the use of such technologies for the visualization of heritage items we can cite, for instance, the on-site augmented and inhabited real time visualization of a Pompeian Thermopolium [PAPAGIANNAKIS05b] [VLAHAKIS03], which was prepared in the frame of the European project LIFPLUS [LIFEPLUS08] as well as the museum guide application targeted at the Guggenheim Museum of contemporary art of Bilbao developed by [ABAWI04]. As a further example that illustrates the employment of such visualization approach, we can also cite the ARCHEOGUIDE project [STRICKER01], which was aimed at providing a three dimensional augmented interactive on-site restitution of the Ancient site of Olympia.

106 Related work: taxonomy and review Digitally augmented movies Digitally augmented movies conceptually share some similarities with renderings, introduced in section , and real time augmented reality simulations, overviewed in section As such, they can be described as the process to supplement filmed footage with computer generated virtual elements that appear to seamlessly coexist in the same space as the real ones. Such approach consists in the visualization of a filmed background upon which have been superimposed a set of pre-computed high definition and high quality renderings that have been synchronized with the real camera motion though a camera tracking or a matchmoving process. These movies could also be considered as non real time augmented reality simulations, where the live video feed is replaced by pre-recorded footage and where the real time rendering of the virtual elements constituting the virtual augmentations has been replaced by a static pre-computed computer generated movie. Visual Consistency and Precision Interactivity Automatism Figure 25: digitally augmented movies, classification of the approach The process that is employed to augment real footage with virtual elements is a longer and a more complex process than a simple matchmoved superimposition of pre-computed images. In real time augmented simulations the three dimensional models are computed on the fly by employing a dynamically changing virtual camera which is synchronized with the motion of the head of the user. The creation of virtual augmentations for a movie, on the other hand, usually involves the computing of several separate renderings for each of the frames, and sometimes for each of the virtual objects that constitute the final animation that is overlaid on top of the real footage. Such renderings can contain specific partial information on the three dimensional models and the virtual scene, such as, for instance, the results of complex lighting simulations, the distribution of material and objet identifiers, the direction toward which the polygons that constitute the models are facing or the distance from the camera at which each object is located. Such approach is commonly referred as multipass rendering, and is usually employed to provide more flexibility and freedom during the compositing and post production phases [WRIGHT06] [KELLY00] [BRINKMANN99] where all the pre-computed renderings are rearranged, modified, color corrected, masked, combined and finally integrated, layer by layer, on top of the real footage.

107 106 Methodological approaches for the protection of cultural heritage in the digital age As a side note, we would like to mention that the compositing processes and the techniques that are used for the creation of augmented movies are not exclusively limited to the methods that allow a real scene to be supplemented with virtual elements, but do encompass also the augmentation of digitally created environments with real elements. Such augmentation is generally achieved by filming the real elements in front of a blue or green screen and by masking them by employing hardware based [ULTIMATTE08] or software based [IMAGICA08] chroma key techniques [ALBUQUERQUE00]. By means of this process a transparency matte, or alpha channel, is extracted and assigned to the image sequences using color, luma or chroma selections: this allows the real elements to be effectively cut out from their colored original backgrounds in order to be superimposed to the digital environments. The creation of a fully featured digitally augmented movie is a process that requires the work of several people over long periods of time and as well as a considerable amount of time and effort to reach completion. Moreover, since the final visual results are standard videos exhibiting the inclusion of digitally created visual effects, which can be projected on a screen, visualized on a television or displayed on a personal computer for the viewer to passively see, their degree of interactivity is therefore extremely limited. However, as it is not necessary to follow strict limitations regarding the maximum computing time used for the rendering of each frame of the animation, complex physical simulations and manual retouches can be applied for the production and for the finalization of the digital augmentations. Hence, analogously to what introduced in section in respect to digitally created renderings, such approach has the potential to produce extremely visually consistent and extremely precise visualizations. Nevertheless, the full expression of such possibilities is most commonly achieved only by the movie industry in the frame of the publishing of blockbuster movies, which may justify the inherent costs of the inclusion of such complex and visually convincing effects by the income they generate at the box office. However, although the visualization of extremely detailed and historically correct augmented environments can be achieved in the frame of such projects, most of the time the visualizations of heritage items that are offered by such high production value movies are usually lacking scientific precision and historical accuracy to be considered as a reliable tool for heritage protection related activities. The simple reason is that commercially exploitable fully featured films are first aiming at entertaining their audiences with breathtaking visual experiences while keeping their production costs as low as possible. Consequently, since the desire to inform, teach, educate or disseminate information relative to heritage items is usually not their priority, as long as the visualized items look sufficiently visually appealing, and as far as the final result gives the illusion of being consistent enough with the depicted historical period while serving the aesthetic and dramatic objectives set by the narrative frame of the movie, no additional resources are generally spent to increase the geometrical precision and the historical fidelity of the virtual elements. As a few examples of digitally augmented high production value movies featuring a visually consistent, but rarely scientifically precise restitution of ancient heritage sites [WARD01] we can mention, for instance, the television series Rome [MILIUS07], which is set during the historical period that witnessed

108 Related work: taxonomy and review 107 the transition from the Roman Republic to the Roman Empire, the movie Gladiator [SCOTT00], which featured a photorealistic digital restitution of the ancient city of Rome and its Coliseum, as allegedly appeared during the second century, or the movie Kingdom of Heaven [SCOTT05], which depicts the ancient Jerusalem at the time of the Crusades. Additionally we can also cite the short demonstrative movies that were created in the frame of the LIFEPLUS project [LIFEPLUS08], which featured selected views of the Pompeii site augmented with historically consistent and scientifically correct virtual humans, and the movie 300 [SNYDER07], in which digital environments were augmented with real actors, filmed on a green screen, to achieve a fictionalized and aesthetically surreal presentation of the battle of Thermopylae Semantically supplemented representations The term semantics is usually employed to describe the relations that exist between things, or between conceptual constructs, as well as to express the relative significance that such elements exhibit for a receiver or a viewer. The process of augmenting a visual representation with a semantic dimension consist in supplementing the graphical data, such as two dimensional photographic images [GEORGOPOULOS03], digital panoramas [VALTOLINA06], or three dimensional virtual models [CALORI04] with a set of structured information in the form of meaningful attributes and concepts that provide a context and a meaning when associated with the visualized items. The semantic descriptions that can be attached to a visual representation can include information about the relations that exist between its constitutive elements as well as between such elements and the real world. A cylinder, for instance, is apprehended by a computer as a set of vertices and faces in a three dimensional space, however it is possible to supplement such description with other data which is related to the function and meaning of such geometrical element. Hence, this cylinder could be meaningfully described as a column exhibiting a specific order, such as Doric, which characterized ancient temples pertaining to a particular historical period, and whose relative positioning inside these buildings is related to specific architectural rules. Additionally, information on the specific parts that constitute a column, which is divided into a shaft, a base and a capital, could also be provided alongside the conceptual associations that are linking these architectural elements to the other structures constituting a temple. A column, for instance, is generally employed to sustain an entablature, which is constituted by an architrave, a frieze and a cornice. Finally, each of these elements could also be augmented with information about the location of the building, the architect that built it, or even with specific data that is directly accessed by the users through the use of interactive hotspots, such as for instance the description and historical details that are related to each of the sculptures eventually present on the pediment of a temple. As such, each of the interconnected elements represented by a given visualization, which are acting as containers holding formal and informal information as well as geometrical and relational data [DONATH03], can become a gateway to access different layers of non graphical information that is related to the visualized heritage property, therefore supplementing the considered object with new dimensions other than its mere visual appearance.

109 108 Methodological approaches for the protection of cultural heritage in the digital age Visual Consistency and Precision Interactivity Automatism Figure 26: semantically supplemented representations, classification of the approach applied to 2D photographs in white, applied to 3D real time visualizations in gray Since computers do not exhibit, as humans do, the capacity to reliably process the semantic properties of raw visual data, the inclusion of structured collections of information and sets of inference rules [BERNERSLEE01], which provide a context and a meaning that is associated with the visualized items, is thus necessary in order to enhance the capacity of computer systems to handle such content. In order to access the digital knowledge embedded into the different elements that constitute the represented item, which requires several layers of heterogeneous interconnected information, it is therefore inevitable to implement specifically designed conceptual descriptive tools, adapted data retrieval and extraction mechanisms, and suitable interfaces to browse and display such information. As such, in order to describe the links that can exist between the different concepts that are connected to the visualized item, as well as to provide a conceptual frame for the metadata that is associated to its constitutive elements, the organization of the content in an ontological hypertextual structure [FALQUET03], that formally defines specific classes of objects and the relations among them, is usually employed. As a result, the graphical representation serves as a base upon which an hyperlinked network of indexed terms and concepts [VANOSSENBRUGGEN01] is implemented, thus offering the possibility to associate a specific meaning to the contents of the visualization while allowing the users to access the data that is related, either directly or indirectly, to the concepts linked to a specific visualized item. In order to provide an efficient way to process, store, reuse, retrieve and exploit such data, the information is usually organized in a structured database which is implements the ontological model describing the specific classes of objects that constitute the content of a considered visualization. The implementation of semantically structured data organized in digital repositories can positively impact the visualization of heritage items since it would offer different classes of users the means to flexibly access, pertinently retrieve, efficiently browse and simply present, depending on their specific requirements and needs, the varied multimedia information that is associated with a specific cultural property. The exploitation of rich semantic models would also allow such users to execute specific searches concerning any conceptually

110 Related work: taxonomy and review 109 related material across entire databases, thus providing such users with the capacity to explore in greater detail the relations that exist between different cultural items [PITZALIS06], and would provide them with the chance to better understand in their entirety the cultural dimensions featured by a specific heritage item. The full integration of database technologies and querying systems with real time virtual reality environments [BERTINO05] would further expand the interactive capabilities exhibited by virtual cultural heritage visualizations while offering their users intuitive and efficient interfacing mechanisms to access the content and the information they need. As such, it is understandable that this approach can benefit several areas related to the protection of cultural heritage. Semantically contextualized data can successfully support activities that range, for instance, from the development of heritage information management tools for museum curators [DOULAVERAKIS05] to the creation of customizable guided tours for tourists visiting open heritage sites [KRENN06] and closed museums [MAZZOLENI03]. Additionally, the implementation of semantically augmented representations can equally constitute a valuable asset for the analysis of the documentation related to architectural heritage [DUDEK04] or for the classification of the material and the information that is employed in the frame of virtual museum projects [FALQUET01] [SRINIVASAN05]. Furthermore, the application of such strategy can also assist the modeling [GRUSSENMEYER99], the reconstruction and the in situ recording [COSMAS02] of the documentation that is collected at an archeological excavation site. Semantic dimensions can be associated with almost any type of virtual representation, as long as they provide the user with a minimum degree of interactivity. Such approach can be combined with two dimensional and three dimensional data, as well as with purely virtual or mixed reality visualizations of heritage properties. Whenever an interactive representation is supplemented with structured semantic descriptions its exhibited degrees of virtuality and interactivity will generally increase at the expense of its degree of automatism, since the introduction of a semantic layer implies the preparation of a larger amount of digital knowledge to describe the considered virtual properties as well as longer development cycles to associate such metadata with the visualized items. As a few examples that integrate semantic descriptions into structured graphical virtual representations of cultural heritage items we can mention, for instance, the SCULPTEUR project [SINCLAIR05], which aimed at assisting museums with the handling of multimedia information by employing semantic web technologies, or the ARCO project [ARCO06], which was targeted at providing curatorial staff with a tool to manage the virtual representations through XML driven cultural object databases. Additionally, we can cite the reconstruction of the temple of Nasium [BUR03], in which the implementation of relational database structure allowed a rational and synthetic organization of the all the items that were collected during the on-site excavation as well as during their anastylosis, and the photogrammetric documentation of the Opera House in Hannover [KARSLI03], in which a database was employed to associate both graphical and non-graphical data to allow the conduction of queries, updates and analysis on the semantically described properties of the model.

111 110 Methodological approaches for the protection of cultural heritage in the digital age 3.3 Modeling strategies for visualizations exhibiting high virtuality In this section, an overview of some of the most diffused techniques that are usually employed to create the content for a visual representation will be briefly presented. However, the considered techniques will be limited to the modeling strategies that are directly related to the preparation of virtual representations of heritage items by means digital visualization technologies. The content creation approaches that will be introduced in the following sections exhibit the capacity to be fully implemented and exploited solely in the frame of the visualization projects that belong to the third and fourth quadrants of the previously introduced taxonomical space. As such, only the digital modeling and the digital representation strategies that can be employed in the frame of computer aided visualization projects that feature a sufficient degree of virtuality will be considered for this overview. To facilitate the presentation of the distinctive qualities that characterize each of the considered digital modeling techniques, as well as to illustrate and emphasize their specific strengths and weaknesses, a conceptual categorization of such approaches, as introduced by figure 27, is proposed. To such end, the taxonomical model developed in section 3.1 has been used as support for the classification of each of the considered visual representation strategies. The localization of each modeling strategy inside this conceptual space accounts for its ability to provide and facilitate, in the frame of a specific visualization approach, the expression of a certain degree of visual consistency, precision, automatism or interactivity. Although, the same restrictions concerning the compatible combinations of dimensions, as introduced in section 3.1, do also apply for the classification of the modeling approaches, the trade-off mechanisms involved in the categorization of such techniques have been implemented in a more simple way. In fact, each given combination of dimensions is visually represented as exerting an absolute constrain in respect to the trade-off they imply. Therefore, each approach can support a given visualization project with the exclusive capacity either to enhance the intensity exhibited by a single dimension or to assist the full expression of a pair of dimensions. Such choice allows for a simplified graphical localization and classification of each of the considered modeling strategies: these are located either on the vertices of the conceptual model or on the midpoints of the segments that delimit the boundaries of the taxonomical space. Since the visual representation strategies introduced in section 3.2 are prevalently located inside such space, the visualizations that exhibit a high virtuality and the modeling techniques that can be used to provide such attempts with a suitable content can be combined in order to highlight their interrelations. Although the considered modeling approaches can be applied to any visualization strategy that is located in the third and fourth quadrants of the classification model defined by the continua of virtuality and interactivity, such techniques are better adapted to some visualization projects rather than others. In fact, the implementation of a modeling strategy that can quickly produce

112 Related work: taxonomy and review 111 simple but highly interactive models in the frame of a visualization project whose goal is to produce highly precise and visually consistent results, at the expense of its overall of interactivity, would be counterproductive, since it would negate the main benefits inherent to the chosen approach. Even though the creation of such content would require a short development cycle, the resulting visualization would exhibit a low degree of interactivity coupled with a virtual representation that would display low degrees of precision and visual consistency, since such dimensions were traded off by the chosen modeling approach. Visual Consistency and Precision Polygonal modeling 2. 3D Scanning 3. Image based modeling 4. VRML modeling 5. Spline surfaces 6. Digital photography Interactivity Automatism Figure 27: modeling strategies, classification of the approaches and areas of affinity To schematize the relations that exist between a chosen visualization approach and a given modeling strategy, we can define around each vertices of the classification model an area of affinity, which is presented as a gray circle in figure 27. Such area encompasses a portion of the taxonomical space which contains the visualization approaches classified in section 3.2 and the modeling techniques that are located along the boundaries of such space. Hence, it is possible to define three main areas that contain three modeling techniques each, three intersecting secondary zones that contain one modeling approach each, and an empty neutral zone which is located near the center of the taxonomical space. Additionally, we can note that each of the defined areas of affinity can be divided in three subsections which are constituted by a primary zone located near a vertex of the model, and by two intersecting secondary parcels that are encompassing a segment midpoint. As such, we can graphically represent the pertinence exhibited by a given modeling strategy in respect to its eventual application in the frame of a particular visualization approach classified inside the taxonomical space. More specifically, a given digital content creation strategy will be more pertinently employed and more suitably exploited in the frame of the visualization approaches that are classified inside its specific affinity subsection, or eventually, but not always, inside its general confidence area. As a concrete example we can consider the creation of a detailed computer generated movie, introduced in section , which is located in the

113 112 Methodological approaches for the protection of cultural heritage in the digital age taxonomical space around point 4. The projects that employ this approach have the potential to exhibit an extremely high geometrical precision and an impressive visual consistency; however they require long development cycles and they offer a very limited interactivity. If we refer to figure 27, we could employ spline surfaces, as a first choice, or scanning and polygonal modeling techniques to develop digital content that supports such type of visualization. Either of those choices would preserve the inherent benefits exhibited by the chosen approach, while the implementation of a modeling strategy located outside the affinity area in which such visualization approach is located would sacrifice both visual consistency and geometrical precision. Inversely, we can consider a real time virtual simulation, overviewed in section and located near point 3 of the model. This approach can support highly interactive content that exhibits fairly good degrees of visual consistency and precision. Thus, for the constitution of the three dimensional models to be included in such simulations, polygonal modeling techniques would be the preferred content creation strategy while image based and VRML modeling techniques could be employed as possible alternatives or as complementary approaches. The implementation of such techniques would, in fact, preserve the high degree of interactivity that characterizes real time simulations, while the adoption of a content creation strategy that resides inside another affinity area would negate such benefit Polygonal modeling By using the term polygonal modeling we refer to the content creation techniques that employ manually crafted structured polygonal elements, which are connected to each other by common vertices, to geometrically approximate the surface of a virtual object. A polygonal mesh is generally constructed by associating a list of points, defined in a three dimensional space, to several grouped sets of segments, each of which describes a sequence of interconnected edges that create a polygonal shape delimited by coplanar vertices. Since a triangular shape ensures that its vertices reside on a common plane, thus allowing the simple and unequivocal definition of a surface normal, which prevents the manifestation shading artifacts, triangles are the most common base element that is used in polygonal modeling for the definition of a geometrical mesh. As a side note, we would like to mention that although triangles are the simplest polygons that can be constructed in a Euclidean space, a polygonal model is nonetheless not required to exclusively rely on such elements for its definition. As such, other polygonal shapes can be used as faces for the models: quads, for instance, are particularly adapted for the creation of subdivision surfaces, and n-sided polygonal faces can appear as a byproduct of the application of semi automatic polygon reduction processes to triangle based models. Although several approaches, ranging from the manual description of the vertices and faces of the virtual objects to the implementation of automatic scanning technologies, can be employed to build a polygonal model, the techniques introduced in this specific section will be limited to the modeling strategies that are commonly used to manually craft and refine complex

114 Related work: taxonomy and review 113 polygonal meshes by means of toolsets widely available in 3D graphics software packages. A fist method that can be used to create polygonal meshes, either alone or in conjunction with the ones that will be introduced in the following paragraphs, is by employing primitive objects, such as spheres, boxes, cones or cylinders for instance. Such primitives can be parametrically modified or manually adjusted and deformed at vertex level to suit the needs of the modeling process. Furthermore, these basic shapes can be connected by means of Boolean operations, such as union, intersection or subtraction, thus allowing the generation of more complex compound volumes that can serve as a base for further combinations. Moreover, compound volumes can also be created by associating a spatial transformation, such as a translation or a rotation, to arbitrary two dimensional shapes, hence offering the possibility to create parametric shapes as, for instance, extruded models, lofted objects or lathed surfaces. A second approach that can be employed to construct complex volumes is by using photographic references of the object whose shape is to be modeled, usually depicting a front and a side view, and by crating and placing the polygonal faces by edge extrusion and face subdivision. The modeling process generally starts by placing a two dimensional plane in one of orthographic views containing the reference images, and by splitting and replicating its edges in an offset position to generate a new contiguous face that is covering a new portion of the targeted surface. Subsequently, the vertices of the newly created polygon are adjusted by employing the second reference image to match in the three dimensional space the surface of the modeled object. Such approach generates strips of polygons, generally composed by quads, since their subdivision produces more regular meshes that triangles, which follow the main construction lines that define the shape of the modeled object. These strips are then stitched together to build the final appearance of the model. By analogy, we may can describe such modeling strategy as a process that is aimed at progressively wrapping the three dimensional volume of the modeled object, inferred from the two dimensional photographic reference projections, with a series connected bands of quad surfaces. A further modeling approach that is commonly adopted to constitute polygonal models, which also relies on four sided polygons as base elements for the definition of their meshes, is referred as box modeling. Such approach employs face subdivisions and face extrusions as main modeling tools. However, the box modeling process, which may, or may not, rely on reference images to assist the modeling phases, usually starts by placing a three dimensional box in the scene. Subsequently, the faces of such primitive are subdivided and extruded, while their vertices are adjusted to better represent the volume of the targeted shape. Such content creation strategy involves the deformation and modification of a base volume which is already closed, instead of the creation of two dimensional surfaces that are conformed and stitched to encompass the volume of the modeled item. Therefore, by analogy, box modeling can also be described as a process akin to sculpting, where the representation of a complex surface is achieved by the progressive refinement of a simple base model with the addition of layered surface details.

115 114 Methodological approaches for the protection of cultural heritage in the digital age In order to enhance the overall visual consistency of the modeled objects, the sole presence of a remarkable level of surface detail and precision is usually not sufficient. Therefore, further steps are generally required to improve the visual impact featured by a polygonal mesh: these operations range from the definition and application of realistic shading and material attributes, to the definition of the physical properties of each modeled item for simulation purposes, to the creation of detailed textures which contain the two dimensional representations of the geometrical features that are deemed negligible in comparison to the scale of the represented item and the scope of the visualization and which need to be applied to the model surfaces. Moreover, since a polygonal mesh can present hidden geometrical problems, such as self intersecting surfaces, overlapping faces, broken edges or stacked vertices, which can cause severe visualization and rendering artifacts, a further time consuming process aimed at cleaning and inspecting the model, as well as aimed at fixing and correcting such geometrical inconsistencies, is usually necessary before the finalization of the content. The application of polygonal modeling techniques to represent three dimensional models offers several benefits. First of all, polygonal representations are particularly adapted to be directly rendered by using scanline or raytrace algorithms, thus allowing even fairly complex and detailed models to exhibit interactive frame rates in a real time simulated environment. Additionally, such approach equally provides an optimal control over the total amount of geometrical detail as well as over the total number of polygons that are used to render the geometrical complexity of the modeled items, which directly impacts their interactive capabilities in a real time environment. As such, polygonal content creation strategies allow the careful preparation and the optimization of 3D meshes that feature structural elements characterized by extremely complex geometrical features while offering the possibility to give them special attention during the modeling phase to ensure an acceptable trade-off between visual accuracy and the performance of the real time simulation [THALMANN07]. On the flipside, a few shortcomings that characterize polygonal modeling strategies are to be mentioned. The foremost disadvantage exhibited by polygonal representations, since their meshes are constituted by discrete elements and not by continuous surfaces, is their incapacity to accurately restitute the appearance of curved structures. In order to achieve a very precise and visually consistent approximation of a curved surface, a polygonal mesh requires the creation of a considerable amount of polygons, thus limiting its capacity to be interactively displayed in a real time environment. Additionally, as a second limitation that is exhibited by purely polygonal models, we can cite the fact that their surfaces are deprived of mathematical descriptions, since they are constituted by a list of points and faces: this precludes, or at best impedes, the possibility to dynamically modify or easily adjust the density and the distribution of their triangulated meshes without affecting the transformations and animations that have been previously applied to their vertices D Scanning Manually crafted meshes obtained through the application of polygonal modeling computer assisted content creation techniques have the ability to create visually appealing, as well as highly interactive, virtual representations. However, the

116 Related work: taxonomy and review 115 amount of geometrical detail that is required by specific visualization projects, as well as their set development cycles, sometimes exceeds the ones that are offered by such strategies. Therefore, in order to quickly constitute very precise and very detailed high fidelity virtual representations, the implementation of automated surface acquisition and reconstruction techniques, through the application of scanning technologies [BLAIS04], is often preferred and widely employed [GUIDI06] whenever the overall interactivity of the final visualization is not considered a critical feature. Scanners can generally be described as devices that are capable to detect and acquire the shape, and eventually the visual appearance, of a real object by sampling and analyzing the object surfaces either by contact, by actively emitting light, radiation or sound to probe the considered object, or by passively detecting the visible ambient or infrared light that is naturally reflected by the item s surfaces, i.e. by employing an image based approach. Since the first category of scanners is generally not well suited to acquire heritage items, due to the fact that a direct physical contact with the objects themselves could compromise their physical integrity, and since passive scanning techniques, which can be referred as image based modeling strategies are covered in section 3.3.3, in this section we will mainly offer an overview of active scanning approaches. Active scanners, either based on time-of-flight computation, triangulation techniques, structured light or modulated light projection, collect three dimensional positional data, end eventually color information, by probing an object s surfaces though the emission of some kind of radiation and by analyzing the response that is provided by such surfaces: this analysis is typically based on the reflectance values of the probed surfaces. The acquired data can be digitally visualized as a point cloud; however, such step requires further processing [BERNARDINI02] in order to be effectively employed to constitute a virtual representation of a given item. As such, the collected positional information is post-processed and converted into a polygonal model or used as a base to extrapolate a parametric surface [MONTI03]. The most commonly adopted approach is to constitute a polygonal mesh though the application of specific surface reconstruction algorithms, whose task is to identify adjacent points and to connect them to generate a continuous polygonal structure that can be textured by employing the surface color data that was eventually extracted during the scanning process. Amid such surface reconstruction tools we can cite, for instance, the Meshlab [MESHLAB08] software jointly developed by the European Networks of Excellence Epoch [EPOCH08] and Aim@Shape [AIMATSHAPE08], which provides mesh cleaning and remeshing filters to process and edit unstructured 3D triangular meshes, as well as a set of tools for editing, cleaning, healing, inspecting, rendering and converting this kind of meshes. Optical scanning devices, such as laser scanners, have the ability to automatically and systemically acquire hundreds of thousands of points per second with a very fine resolution, as illustrated by the results achieved in frame of the Digital Michelangelo Project [LEVOY00] where some datasets related to the digital reconstructed model of the statue of David were reaching a 2 billion polygons count and a resolution of about 0.25 millimeters. As a side note, we

117 116 Methodological approaches for the protection of cultural heritage in the digital age would like to mention that although the scanning process is usually performed at a resolution which is generally proportional to the object size, hence employing a lower resolution for large items, such as buildings or statues, and a higher one for small detailed artifacts, it is however possible to otherwise acquire large surfaces with an extremely accurate level of detail, as it is the case for the physical model of the imperial Rome addressed in [GUIDI05] which features the implementation of close range high resolution and low noise scanning techniques. Furthermore, due to their varying size and weight, scanning devices are not exclusively confined to stationary and fixed positions but can be integrated in mobile and even airborne systems, as it is illustrated by the digital reconstruction of the Bayon temple in Cambodia [BANNO07] which was carried out by using a moving range sensor attached to a balloon. Since such shape acquisition approaches allow extreme levels of precision, flexibility and fidelity, they can be employed as an effective way to constitute high quality records of cultural properties that can be exploited for varied heritage protection related activities, such as for archival, display, research, conservation, replication and even for virtual reality applications [TAYLOR03]. Moreover, the application of such techniques can range from the accurate three dimensional representation of small objects, such as library materials [BROWN00], to the documentation of large spaces, such as the terrestrial scanning sessions that were performed in the interior of Castle of Neuschwanstein presented in [ABMAYR05]. Automatic active surface acquisition strategies and scanning devices, despite the excellent degree of geometrical precision and short development cycles they offer, do however present, aside from the fact that they are generally quite expensive, a certain number of shortcomings compared to other digital content creation approaches. A first element that should be considered whenever performing the digital acquisition of an item by means of scanning technologies is the optically cooperative nature of the surfaces of the targeted object. While most surfaces can be probed without the creation of significant errors in the recorded data, the scanning of very dark, shiny, transparent or translucent objects can generate incorrect positional data due to the optical effects they produce, such as subsurface scattering and refraction phenomena as well as high light absorption or high reflection surface properties. Although some techniques have been developed to allow the correct, but generally much slower, active acquisition of optically uncooperative surfaces, such as for instance the implementation of laser radars [RICCI03], further aspects related to the technical specifications of the employed devices [BOEHLER03], such as their angular accuracy, their resolution, their optimal acquisition distance as well as the amount of noise and edge effects they generate in the recovered datasets, should be considered to ensure that the acquired data allows for the three dimensional reconstruction of an exploitable mesh. A second element that requires attention when a data acquisition is performed by employing such techniques is the fact that more often than not several concurrent or subsequent sweeps are necessary to recover all necessary positional data that is needed for the geometrical representation of the desired object. Although the acquisition of smaller and simpler items can be achieved by means of a few datasets, probing complex surfaces and scanning large spaces,

118 Related work: taxonomy and review 117 such as heritage edifices, requires numerous surface measures. To partially overcome the inevitable geometrical occlusions that characterize the inherent nature of such complex volumes, several acquisition stations are necessary to obtain overlapping datasets that describe the totality of the surfaces of the targeted object. Subsequently, the recorded positional data is combined and filtered by means of specific surface reconstruction software to generate a continuous mesh. However, triangulation problems and noise artifacts, coupled with holes caused by missing data, usually impede a fully automatic generation of such meshes and the manual user intervention is often required to oversee and direct the filtering and the post processing of the recovered datasets. A third element which should be noted is that digital scanning strategies generally produce very dense meshes, thus disallowing their direct exploitation for any interactive purpose unless properly cleaned and decimated, which however would imply a decreased precision for the generated model, therefore negating the main benefit of such approach. Furthermore, it has to be stressed that unlike other content creation strategies, which either allow or facilitate the conduction of digital virtual restitution experiences targeting the hypothetical past appearance of a considered cultural item, scanning devices are limited to the acquisition and documentation of the shape and the still existing features exhibited by the investigated item. Finally, while referring to [GUARNIERI04] and [BERALDIN04], we would like to mention that active scanning techniques and passive image based modeling strategies should not be seen as competitive approaches, but they rather constitute complementary methods that can be employed concurrently to benefit from their respective strengths. Thus, passive scanning techniques can be used as a fast and low cost approach to quickly create a simplified model of the object of interest, while active scanning devices can be employed whenever complex surface details require a precise documentation by means of the direct acquisition of large amount of three dimensional positional data. As an example of such integrated approach we can cite, for instance, the visualization of the Abbey of Pomposa, near Ferrara, and the digital representation of the Scrovegni Chapel, in Padova, as presented in [ELHAKIM03], as well as the three dimensional model of Pinchango Alto, in Peru, as introduced in [KARSTEN07], all of which were created through the synergetic combination of both image and range data Image based modeling A further content creation approach that can be employed to create a polygonal representation of a heritage property is represented by Image Based Modeling strategies [REMONDINO06] [FARINELLA06]. These techniques can be used to create a 3D model by means of a series of reference photographs that are employed either to assist the modeling of a given object or to generate semi automatically a complete triangulated mesh from the analysis of such photographic references. In order to create an exploitable three dimensional space that allows for the semi automatic definition of an object s geometry, two dimensional images which feature several overlapping elements are used as sources for the Image Based

119 118 Methodological approaches for the protection of cultural heritage in the digital age Modeling process. Several pictures which portray the targeted item from under different points of view are necessary to allow the manual definition and the manual placement of several locator points, each of which corresponds to a specific feature identifiable in at least three different photographs. Such locators are used as a base to estimate the focal length of the cameras that correspond to each provided image, and are employed to build a three dimensional environment that is aligned with the reference photographs. By using the generated 3D space and the recovered three dimensional positions of the manually defined locators, the target object is modeled, usually by employing primitives such as boxes or cylinders, or by employing simple manually defined geometries, while using such locator points as modeling references for the placement of the vertices and the faces of the modeled meshes. Since each reference image is assigned to a virtual camera that corresponds to the recovered source image s viewpoint, it is possible to utilize such pictures to texture the surfaces of the three dimensional model by employing them as texture projectors. As a practical example that illustrates such content creation strategy we can refer to the reconstruction of some of the decorative architectural elements of the Greek temple in Messene in the frame of the CREATE project [LOSCOS03] [DRETTAKIS04], which were realized by means of the specialized image based modeling software ImageModeler [AUTODESK08b]. Although the use of photographic references remains a widely diffused approach, image based modeling techniques can be implemented by using video footage as modeling source, since it could be considered as a sequence of closely taken contiguous pictures. However, as it would be impractical to exploit such large collections of images for the manual placement of locators, camera tracking techniques can be used instead to preprocess the reference videos. Since, the automatic 3D image based feature tracking process performed by matchmoving software provides as a byproduct a three dimensional space which is aligned with the recovered camera motion, it is therefore possible to exploit the extracted feature points, represented as a point cloud, as locators for the image based modeling process [VANDENHENGEL07]. Furthermore, if the extracted points are not sufficient or if their positions are not adapted to describe the placement of the vertices of the primitives to be used to represent the scene, it is still possible to exploit the reconstructed three dimensional space to manually define specific locators that can help the definition of the geometry of the represented objects. The 3D location of these user selected points can be determined, if specified on at least two different frames of the source video, by analyzing their positions relative to the available feature points that were generated by the tracking process. Moreover, textures can also be extracted from the video by using different positions of the virtual camera as camera mapping projectors. A second strategy that equally relies on image references to constitute a three dimensional representation of the objects that need to be modeled is represented by automatic photogrammetric reconstructions, which can be classified as passive image based scanning techniques. The first step that is required in order to succeed in the creation of such models is to shoot a set of photographs taken from a fixed distance around the considered object. Then the

120 Related work: taxonomy and review 119 acquired two dimensional data is used to calibrate the camera as well as to automatically extract a series of depth maps, or range maps, for each of the input images [VERGAUWEN06]. To assist the process, the pictures can also be manually masked to enable a faster and more reliable determination of the object shape through feature comparison and shape from silhouette techniques: the shape of the object can then be recovered from the identified contours by means of volume intersection algorithms [PATIAS06]. The photogrammetric reconstruction web service developed by the European Network of Excellence Epoch, for instance, offers the possibility to remotely process a set of uploaded images for depth map computation and to locally select a specific photograph, once the results have been downloaded, as a reference viewpoint for the creation of a view dependant 3D model while surface textures, which have been extracted from such pictures, are projected and applied on the faces of the generated mesh. However, to build a complete view independent mesh of the considered object, several depth maps are needed to recover the depth of all the feature points, contours, volumes and surfaces that can be identified across all the images in a set. To complete such task, surface reconstruction software, akin to the tools that are employed to triangulate the meshes of laser scanned surfaces, is generally employed. For instance, the Meshlab software previously introduced in section 3.2.2, also provides the generation of a shaded point cloud from a set of depth maps, as well as its subsequent conversion into a triangle based mesh that can be decimated, simplified, filtered and cleaned to produce an interactively usable result. As a further example of the implementation of such techniques, we can also cite, for instance, the automatic surface reconstruction process, based on feature lines and feature points extraction methods presented in [SORMANN04], which were employed to build an image based modeled textured mesh of the Josefsplatz in Vienna. While purely automatic image based modeling approaches can often produce unexploitable results that present large holes and considerable surface irregularities due to the lack of suitable source data, semi automatic approaches still remain one of the most inexpensive, flexible, quick and widely adopted strategies to produce simple geometrical models that can be displayed and animated in a real time environment. However, such high degree of interactivity and such short development cycles forcibly imply a decreased geometrical precision and a weakened visual consistency in the three dimensional models that are produced by means of such content creation strategies VRML modeling To create a polygonal representation of a given item it is not strictly necessary to rely on photographic references, automatic acquisition techniques or sophisticated toolsets. The implementation of these strategies can benefit the visual consistency, geometrical precision, as well as the length of the development cycles necessary to complete a three dimensional model; however, it is possible to adopt a completely manual approach to define the meshes that are used to the represent such objects.

121 120 Methodological approaches for the protection of cultural heritage in the digital age The Virtual Reality Modeling Language (VRML), formerly known as Virtual Reality Markup Language [PARISI94], exhibits a declarative structure that can be stored into a normal text file. Such files, which were originally developed for the Web, and thus intended to be parsed and visualized via web browsers equipped with specifically adapted plug-ins, allow the textual description of three dimensional vector data. The VRML syntax, exemplified by figure 28, allows the manual specification of the vertices, edges and polygons that constitute a polygonal mesh, as well the application of surface colors, textures and material properties, such as shininess and transparency. Furthermore, it equally offers the possibility to integrate into the visualized worlds hyperlinked medias, animated objects, light sources and event triggers, as well as it supports the inclusion of scripted behaviors through the layering, for instance, of JavaScript program code. Shape { appearance Appearance { material Material {diffusecolor ambientintensity 0.55 }} geometry IndexedFaceSet { coord Coordinate { point [ 0 0 0, 0 0 1, 1 0 1, 1 0 0, , ]} coordindex [ 3,2,1,0,-1, 1,4,0,-1, 1,2,4,-1, 2,3,4,-1, 3,0,4,-1, ]} } Figure 28: example of VRML code specifying the vertices and faces, 4 triangles and 1 quad, of a simple red pyramid Most 3D graphics and CAD software packages still include the option to export the modeled meshes into VRML format, and although the manual definition of 3D meshes offers the possibility to create very clean, light and simple models that can easily be visualized in three dimensional interactive real time environments, such content creation strategy however requires long development cycles and a considerable amount of time and effort to achieve completion. Additionally, since such code has to be written from scratch and since it requires several adjustments whenever it is generated from inside a

122 Related work: taxonomy and review 121 modeling application, the visualized models generally exhibit, for practical reasons, a limited geometrical complexity, thus precluding the possibility to manually define very detailed meshes that exhibit high degrees of visual consistency and precision As a concrete example that illustrates the practical possibilities offered by VRML modeling approaches, as well as the inherent visual limitations implied by the adoption of such strategy for the visualization of heritage items, we can refer, for instance, to the recreation of the ceremonial center of the ancient Aztec city of Tenochtitlan built by Delle Maxwell on the basis of the AutoCAD models that were provided by Bob Galbraith as presented in [HARTMAN97]. Finally, VRML browsers and VRML visualizations, despite the success they experienced during the nineties, have become nowadays obsolete, since they do not natively support recent real time rendering techniques, such as pixel shaders, normal mapping or soft shadowing, which contribute to the overall visual appeal of the visualized items. While such modeling strategy could still be encountered for the visualization of simple models in the frame, for instance, of online virtual museum projects, pure VRML modeling approaches have nonetheless been progressively abandoned for the visualization of complex heritage items in favor of more flexible and performing solutions that generally rely on commercial or open source extensible rendering engines, such as, for instance, the Unreal [DELEON00], Torque [WYELD07], or Open Scene Graph [PESCARIN05] renderers Spline surfaces A further approach that can be employed to geometrically create a three dimensional visual representation of a heritage item is the use of parametric and non-uniform rational B-spline surfaces [PIEGL97] [FARIN99]. The pioneering development of such type of representation was initiated in the 1950 s by Pierre Bézier and Paul de Casteljau, who respectively worked for the car manufacturing companies Renault and Citroën in France. Their work was originally motivated by the need to provide car designers with a set of tools that would allow the predictable, flexible and precise creation and manipulation of freeform surfaces, such as the ones used in car and ship hull design. Since spline surfaces provide a compact and efficient representation of complex geometrical shapes, while still allowing a certain degree of ease in their manipulation and modification, they were rapidly adopted for other design purposes and are now widely available inside most computer aided drawing software packages. Spline surfaces can be described as piecewise polynomial surfaces which are built by smoothly connecting several surface patches together, and whose shape is constrained and defined by a set of control points that are employed to interpolate or approximate their surfaces [CATCH08]. Several families of splines have been proposed and the main difference between them resides in the polynomial basis that is employed to interpolate or approximate the given control points: among such families we can cite the Bézier curves, the B-splines and the NURBS surfaces. Due to their continuous representation, spline surfaces are best suited to visualize curved elements since they do not rely on discrete polygonal primitives, such as triangles or quads, to explicitly represent the

123 122 Methodological approaches for the protection of cultural heritage in the digital age surfaces of the visualized objects. Their use can therefore provide an excellent degree of visual consistency and geometrical precision. Nevertheless, it is important to stress that despite the relative ease that the manipulation of a single spline patch offers, the representation of complex surfaces, such as the ones featured by cultural artifacts, requires the manual definition and careful assembling of several patches in order to constitute the final shape of the visualized item, which can prove to be a slow and delicate process. Moreover, unlike polygonal modeling strategies and polygon based representations, spline surfaces are not well suited for real time applications. This limitation is mainly due to the fact that common rendering engines are usually required to triangulate at render time the spline surfaces by converting them in a discrete triangle based representation in order to compute their shading based on the normals of each surface element. To perform such triangulation several methods can be employed, among which we can mention the ones based on curvature and spatial optimization as well as the ones based on the Delaunay approach [LEE80]. It is nevertheless clear that whenever a model exhibits a high geometrical complexity, its corresponding high resolution polygonal representation would negatively impact the level of interactivity featured by its real time visualization, since more time would be required for the triangulation process. In order to alleviate such issue it would be conceivable either to bypass the real time triangulation process by rendering directly the spline surfaces or to preprocess the triangulation offline before feeding the model to the real time renderer. Both solutions present certain flaws that make their implementation impractical. Real time raytracing engines, for instance, could undoubtedly provide an extended set of very interesting features and possibilities that would enhance the visual impact currently offered by standard engines [OWEN99]. Nonetheless, the direct rendering and shading of spline surfaces by means of real time raytracing algorithms is still an open research field. Available real time raytracing strategies [SIGGRAPH05] do generally require large amount of memory and, since they do not fully benefit from currently existing hardware acceleration solutions, they intensively rely on raw computing power: in such circumstances CPU and GPU parallelization as well as cluster based approaches are often exploited to provide satisfactory performances. Even though specific techniques are emerging to address the direct real time rendering of spline surfaces [ABERT06], currently available approaches, due to the lack of sufficient computing power, do not yet allow for the interactive visualization of complex spline based models on average consumer range computers. A second strategy that could be considered to reduce the overall rendering time would be to triangulate the spline surfaces offline. In order to retain the visual accuracy of the original models and restitute the details and appearance of their curved surfaces, such approach would imply the creation of high polygon meshes whenever complex geometries were to be visualized. Since three dimensional models that exhibit a high polygon count would negatively impact the a real time visualization by degreasing its level of interactivity, such strategy would be appropriate only when used to generate simple optimized versions of the original models, such as the ones introduced in section 3.3.1, thus negating

124 Related work: taxonomy and review 123 one of the main advantages of spline surface representations. In this respect, spline representations are generally best suited for offline visualization purposes, such as the ones overviewed in section , where their capacity to produce highly detailed renderings would be fully exploited. As a side note, we would like to mention that while the option to extrapolate or interpolate polygonal surfaces from a spline model does exist, the inverse operation is also viable by means of subdivision techniques. Such operation creates a smooth high resolution polygonal model by iteratively subdividing each of the polygons that define a starting low resolution mesh [ZORIN00] in order to approximate the shape that the object would exhibit if it were represented by spline surfaces. One of the advantages of subdivision surfaces is that each of the subdivision levels remains accessible as a set of control points that can be used to modify and adjust the shape that results from any subsequent subdivision, thus allowing the direct manipulation of the final surface. However, while this approach is often invaluable for the rapid creation of high resolution models that are used for offline visualization purposes, its exploitation for real time interactive purposes remains impractical due to the geometrical complexity that is exhibited by the generated surfaces Digital photography Photographic references, as illustrated by a few of the selected approaches that were previously introduced, can assist and support the geometrical modeling of a heritage property in a three dimensional space. However, two dimensional images can as well constitute per se the core element that is employed for the visualization of a cultural object. Amid the content creation strategies that mainly rely on digital photographic material, or more generically two dimensional digital representations, we can cite digital catalogs, introduced in section , where such data is presented in its simplest form, and digital panoramas, overviewed in section , were the images are mapped on a simple geometry that can be rotated to give the illusion that the user is changing his point of view inside the modeled volume. Since such techniques have already been discussed, in this particular section we will introduce a few additional content creation strategies that are primarily based on the exploitation of two dimensional digital data. Another approach that fits this category is represented by camera mapping techniques. Such method can be considered as a particular subset of image based modeling strategies. By using such approach it is however not possible to produce as a final result a fully three dimensional model of the visualized items, since simple modeled shapes are employed only as a support for the projection of two dimensional images. More specifically, a camera mapping representation supplements a two dimensional representation, such as a photograph, a rendering or a matte painting, with the illusion of a parallax effect that would offer the impression that the visualized picture contains three dimensional modeled elements. In order to give the impression that a third dimension exists in the scene, simple geometrical volumes, such as boxes and planes, are created by means of a 3D graphics software package to constitute a very rough, and usually incomplete, representation of the scene while employing the considered picture as a reference for the placement of such shapes.

125 124 Methodological approaches for the protection of cultural heritage in the digital age Once such geometrical elements have been placed correctly, the two dimensional image is projected from the camera point of view that was used to model the scene, and it is locked as a texture on the surfaces of three dimensional objects. Finally, by employing a second virtual camera, which can only execute limited traveling and panning movements, it is possible to produce, to a certain extent, a convincing parallax effect in the rendered scene, since different parts of the reference image are mapped on different planes that are placed at a varying distance from the second camera. However, for the illusion of depth to stay visually consistent, the camera movements must be confined to a limited range, since large trajectories can reveal parts of the model that are not present in the reference picture. To alleviate such problem, it is current for camera mapping visualizations that must be applied to large camera movements to use several reference images to texture project separate parts of the three dimensional models. These images generally depict different points of view of the scene and are digitally retouched to remove occluding objects: this strategy then allows for a greater flexibility in the animation of the second virtual camera. Although it could be possible to apply such approach to real time visualizations, camera mapping techniques, due to their inherent restrictions and limited interactivity, are however mainly employed in the movie industry to quickly build and animate digital backgrounds whenever an abundance of geometrical detail and precision are not required. A further content creation strategy that primarily relies on the visualization of two dimensional images to visualize a given object or site is represented by the organization of large collections of partially overlapping photographs in a three dimensional space. Such approach, which is akin to digital panoramas, employs camera tracking technologies, feature extraction techniques and comparison strategies to find in a database similar pictures to be presented to the user inside a 3D environment. Feature points are used to place and align the collected pictures in relation to each other by fitting their projection planes and by distorting their shape in a reconstructed three dimensional environment rather than stitching them together in a two dimensional space to produce a single panoramic image. However, the ability to freely navigate the reconstructed space, as well as the overall degree of interactivity that is exhibited by such approach, is relatively limited. To illustrate such shortcomings, we can refer the tech demo of the San Marco Square in Venice based on the Photosynth prototype software developed by Microsoft Labs [PHOTOSYNTH07], where the user s freedom of movement and the size of the modeled space were clearly restricted by the available pictures in the database. This particular content creation strategy offers the ability to present a photo album, or a digital catalog, in a three dimensional environment while visualizing the pictures one after the other and while allowing the user to navigate, from one image to another by following their spatial relation. The Photosynth software, for instance, allows the user to walk or fly through a scene to see the photos from any angle, to seamlessly zoom in or out of a photo, to find similar photos to the ones that are currently visualized as well as to see where such images were taken in relation to each another. As a complementary side note, it is important to stress that such method does not exhibit the capacity to generate a fully three dimensional representation of the visualized items, since only the extracted

126 Related work: taxonomy and review 125 feature points, which are represented as a point cloud, are placed in the reconstructed 3D environment in order to assist the alignment of the planes upon which the pictures are mapped for display. 3.4 Summary In this chapter, a general classification of the different strategies that can be employed to constitute a visual representation of a tangible cultural heritage item, as well as a detailed analysis of specifically chosen techniques, has been presented and discussed. In order to support the creation of a coherent and sufficiently comprehensive taxonomy that allows the classification of all the diverse approaches that can pertinently be employed for such task, specific conceptual dimensions and axes, namely the continua of virtuality, interactivity, precision, visual consistency and automatism, have been defined and analyzed. Such continuous dimensions have been employed as a base to propose three complementary taxonomical spaces whose aim is to provide an insight into the respective advantages and disadvantages of the considered visualization strategies, as well as to offer a conceptual tool that can highlight the situations in which certain technological solutions can best serve specific user requirements. 1. Restitution Drawings 2. Augmented Pictures 3. Scale Models Interactivity Physical Reconstructions 5. Interactive Scale Models 6. Live Experiments 7. Renderings 8. Digital Catalogs 9. Digital panoramas 10. Real time VR simulations 3 1 1a 2 Virtuality Stereoscopic visualizations 12. Videogames 13. Real time AR simulations 14. Digitally augmented movies 15. Semantically supplemented 2D photographs 15a. Semantically supplemented 3D models Figure 29: general overview of the considered visualization strategies as classified using the continuum of interactivity and the continuum of virtually as axes to define a two dimensional taxonomical space. The approaches relying on digital technologies are highlighted by the gray box The first classification model that has been discussed is a two dimensional taxonomical space defined by the association of the continuum of virtuality with the continuum of interactivity. Such model, as summarized by figure 29, has

127 126 Methodological approaches for the protection of cultural heritage in the digital age been divided into four distinct quadrants which have been exploited to characterize the visualization approaches that exclusively rely on traditional techniques as well as the ones that employ modern digital technologies. For each considered quadrant, as well as for some special cases, a few exemplifying strategies and practical representative implementations have been presented to demonstrate that it is possible, by analogy, to fit in the proposed conceptual model other specific approaches that were not explicitly mentioned here. The second classification model that has been introduced, as illustrated by figure 30, combines in a multi dimensional taxonomical space the continua of visual consistency, precision, interactivity and automatism. Such model has been employed to highlight the specific strengths and weaknesses of each of the retained approaches in respect to each other, as well as the inherent tradeoff mechanisms that have to be considered while implementing such visualization strategies. Interactivity Visual Consistency and Precision a 1. Restitution Drawings 2. Augmented Pictures 3. Scale Models 4. Physical Reconstructions 5. Interactive Scale Models 6. Live Experiments Automatism Figure 30: general overview of the visualization strategies which rely on traditional techniques as classified using the continuum of interactivity, precision, visual consistency and automatism as axes to define a multi dimensional taxonomical space Finally, a selection of commonly employed techniques and the content creation strategies that rely on modern visualization technologies to prepare the virtual representation of a heritage item have been introduced and analyzed. Subsequently, our multidimensional taxonomical model has been extended with the inclusion of such elements to allow a more detailed classification of the visualization approaches that rely on such strategies.

128 Related work: taxonomy and review 127 A Visual Consistency and Precision E a B A. Polygonal modeling B. 3D Scanning C. Image based modeling D. VRML modeling E. Spline surfaces F. Digital photography Interactivity D 13 C 8 9 F Automatism 7. Renderings 8. Digital Catalogs 9. Digital panoramas 10. Real time VR simulations 11. Stereoscopic visualizations 12. Videogames 13. Real time AR simulations 14. Digitally augmented movies 15. Semantically supplemented 2D photographs 15a. Semantically supplemented 3D models Figure 31: general overview of the visualization strategies which rely on digital visualization technologies, as classified using the previously employed multi dimensional taxonomical space, and their position inside the confidence areas defined by the considered content creation approaches This extended taxonomical space, depicted in figure 31, has then been used to summarize, to schematize and to highlight the relations that exist between a specific digital visualization approach and a given content creation strategy, as well as to emphasize the pertinence of such approaches in respect to their application in the frame of a particular visualization project.

129 128 Methodological approaches for the protection of cultural heritage in the digital age

130 Methods 129 Chapter 4. Methods Using as a base the conceptual framework that has been developed in chapter 2 and chapter 3, this section will offer a summary of the methods that have been employed to carry out the analysis that will be presented in chapter 5 and whose objective is to offer a complementary insight on the practical implementations of our theoretical models. In order structure our experimental assessment, a set of research questions will be defined alongside the identification of the main hypotheses and assumptions that support our study. Then, an outline of the process that has been employed to construct the tools that have been used for the data gathering phase will be presented. More specifically, an overview of the creation of the questionnaire that has been employed to collect the information that has been processed and the structure of the interviews that have been conducted to support our analysis will be offered. Finally, a few considerations on the statistical tools have been be used, on the selected population sample and on the expected results will be discussed. 4.1 Research question As previously discussed in sections 1.5, 1.6 and 1.7, in which the initial problem statement was first introduced alongside the presentation of the main objectives of our work, we seek to identify for each of the main end user groups that are active in the field of cultural heritage protection the elements that minimize their negative response and maximize their positive one to any given visualization attempt that targets a heritage item. Pursuing such goal should allow the identification of practical guidelines for the preparation of virtual heritage applications that match the expectations and needs of each of such user groups. Furthermore, we aim at identifying whether or not, and to which extent, a convergence exists between those groups, hence allowing the identification of common tendencies that could be prioritized whenever a virtual heritage application is intended to be used by more than one of such groups. In order to offer a coherent structure to our work, as well as to define the conceptual boundaries that are necessary to conduct our experimental assessment, such general objective should be deconstructed into to several simpler questions that will be employed as a guideline for our analysis. As such, the main research questions that have been selected to be used as a frame of reference for our study are as follows: o Which are the priorities of each end user group in respect to their field of o activity? Which are the needs and expectations of each user group in respect to a visualization strategy that targets a heritage item? o How is perceived by each of the groups the usefulness of a given visualization approach? o Which are the elements that constitute a successful virtual cultural heritage application from each of the groups points of view? o o These elements differ from group to group? If yes, to which extent? Which visualization approach is the most appropriate in order to satisfy the specific needs of a given end user group?

131 130 Methodological approaches for the protection of cultural heritage in the digital age o Is there an optimum balance that could simultaneously satisfy the minimum requirements, expectations and needs of each of the groups? By answering those questions, and by employing them as a focus for our experimental assessment, it should be possible to find a solution to our main predicament concerning the identification of general guidelines, based on the common tendencies that can be identified amongst the considered groups, for the creation of multipurpose virtual heritage applications. Although it should be likely, a priori, that most of the research questions that define of our main problem statement should find a satisfactory answer, it is interesting to note, in relation to the last question stated in our list, that it could be possible for our analysis not to provide sufficient indications to highlight a set of converging traits among the studied groups. If such case should present itself, the eventual absence of general common tendencies should nevertheless provide a valuable indication concerning the necessity to treat the analyzed groups as separate entities that exhibit mutually exclusive needs and requirement, thus leading to the elaboration of separate guidelines for each one of them. 4.2 Area of investigation In order to create clear research boundaries for our study, it is necessary to narrow the scope of our experimental assessment. It would otherwise be unlikely to identify any meaningful result if the area of investigation of our analysis indiscriminately encompasses all heritage items, regardless of their type, inherent nature and tangibility, as well as all possible applications that digital technologies can offer to assist the efforts that aim at their protection. Consequently, amidst all the available exploratory directions and research angles that are related to such general area of study, the option to exclusively limit the scope of our work to visualization technologies that target human tangible cultural heritage items, as introduced by the classification presented in section 2.2.5, was preferred. As such, it is inevitable that this thesis presents some inherent limitations. Specific topics, such as the role that digital applications can play in strengthening the activities that are related to the protection of natural or intangible heritage and the contribution that other approaches that are not exclusively based on visual information, such as audio recordings or tactile feedback, could bring to the field of cultural heritage have in fact not been considered as central elements for this study. Therefore, we remain aware that such research directions do raise pertinent questions as well as they do introduce interesting perspectives for the application of digital technologies in the field of heritage protection, however, such topics fall outside the particular scope of our work. 4.3 Experimental assessment The taxonomical models previously introduced in section 3.1.6, which were later extended in sections 3.3 and summarized by figure 29, figure 30 and figure 31 as presented in section 3.4, could provide an answer to most of the research

132 Methods 131 questions that have been stated in section 4.1; however, such conceptual framework remains a theoretical construct and a conceptual tool for the analysis of the role played by visualization technologies in the frame of heritage related applications. Therefore, it would be interesting and meaningful to try to test the validity of such models by confronting them to practical working situations. To achieve such goal, it would be foreseeable to supply a specific interest group with a digitally created visual representation of a heritage item they are interested in and which was developed by employing the same techniques and approaches that are deemed optimal by the aforementioned theoretical framework. Their reactions regarding the suitability of such visual representation to their specific subjectively perceived needs and requirement could then be empirically collected. An analysis based on the presence or the absence of an eventual deviation from the predicted outcome, based on the group s satisfaction, would provide significant information regarding the adequacy of our model to a real world situation. Alternatively, it could offer an indication concerning an eventual inadequacy of the users perception regarding their own real needs, due to an overestimation of their expectations or an underestimation of the real possibilities offered by the currently available visualization techniques. A further direction of study that could provide interesting information would be to directly analyze, either quantitatively or qualitatively, the subjective preferences, needs and requirements that the different user groups exhibit towards the visual representation of an heritage item, by employing as a base the elements and dimensions that have been identified by our conceptual framework and by transposing, for instance, such elements into indicators in a questionnaire. Such an analysis could provide the necessary experimental data that could lead to the identification for each user group of the elements that minimize their negative response and maximize their positive response to a proposed visualization approach, thus allowing the validation of the theoretical guidelines suggested by our conceptual framework. Moreover, it would permit the identification of the main elements that the different user groups consider primarily responsible for a successful representation of a heritage item. Since the first considered option would be impractical and cost ineffective, due to the need to develop several ad hoc computer applications for each of the considered user groups to test their individual reactions, and since the main objective of this study is to provide some directions and ideas on how to develop virtual heritage applications that are better adapted to the specific needs of a targeted end user group, the second option was preferred. Moreover, the information extracted by means of a survey, would allow the integration of empirical data inside the previously proposed conceptual models. Thus, it would be possible to add a new complementary layer to our framework, which would account of the positioning of the considered groups inside the taxonomical spaces, and which would offer a clear and formalized view of their preferences, needs and requirements Definition of the targeted groups of interest In order to conduct our experimental analysis, the definition of clearly identifiable target groups, whose preferences regarding the application of visualization

133 132 Methodological approaches for the protection of cultural heritage in the digital age techniques to heritage items will be assessed in the frame of our study, is a necessary step. Since we are mainly interested in the groups that are directly linked to the activities that participate in the protection of cultural heritage, we can refer to the actors and their intervention areas that were previously defined in section 2.3 and 2.4 and introduced by figure 3. These actors, as well as the particular dimension of the definition of protection of cultural heritage to which they are connected, are summarized by table 3. Actors participating in cultural heritage protection activities Dimensions linked to the protection of cultural heritage Target groups considered for the experimental assessment Content Holders Field Personnel Cultural Intermediaries Investigation + Expertise Conservation + Defense Documentation + Dissemination Archeologists Historians Civil Protection Restoration technicians Teachers Museums Content Consumers targets for the Dissemination General Public Digital Intermediaries technological mediators Computer scientists Computer Graphic artists Table 3: actors participating in cultural heritage protection activities, dimensions of the protection of cultural heritage to which they are mainly linked, and groups selected for the experimental surveys Furthermore, for each of the categories of actors linked to the previously identified dimensions and activities, we selected two specific user groups to be targeted by our experimental assessment Surveys In order to gather de necessary data that is required for our analysis, two different strategies have been pursued in the preparation of the experimental assessment that targets the selected groups of actors. First, the choice to use a questionnaire, available as appendix 2, has been selected in order to study the specific response that each of the considered groups exhibits regarding the use of visualization technologies whenever they are applied to their own areas of activity. Such study should provide sufficient information to establish a general profile of the tendencies that are displayed by each considered group, therefore allowing the highlighting of their specific preferences and requirements regarding a virtual heritage application. By considering their needs and expectations, such analysis should provide the necessary data to empirically establish which visualization strategies would be the more appropriate for a visualization project that targets such groups. In order

134 Methods 133 to carry out such survey, a multiple choice questionnaire that contains sixty seven questions (details of the design of the questionnaire are provided in section 4.6) has been distributed to a selected population sample comprised of ten to twenty individuals that were issued from each of the target groups mentioned in table 3. However, due to practical reasons, whenever two groups were available as representatives for a class of actors, only the first one was retained as participant to the survey. As such, for instance, for the cultural intermediaries category, only a group of teachers was considered as sample for the assessment. The second approach that has been employed to assist our study is a survey in the form of a series of interviews (their design and protocol are detailed in section 4.7), which were targeting specific persons capable of offering an informative, objective and representative point of view on their own specific areas of professional activity. In this case, one to two people that were issued from each of the groups considered for the experimental survey have been interviewed in order to assess their preferences, needs, expectations and requirements concerning the use of visualization technologies in their particular field of expertise. 4.4 Assumptions The main assumption upon which we base our surveys is related to the size of the population sample that has been selected as target for the experimental assessment. It is in fact obvious, under normal circumstances, that a study conducted on a randomly chosen sample of such a limited size would be unlikely to provide significant and reliable indications that could be generalized to the whole population targeted by our analysis. Nonetheless, since the actual sample has not been randomly selected, and since each of the participants to the experimental surveys does belong to a clearly defined professional category that is active in the field of heritage protection, we believe that the information that can be extracted by means of such a questionnaire could be sufficiently consistent to provide acceptable results. However, for such data to be exploitable, we assume that each of the surveys participants displays a sufficiently clear understanding and a sufficient awareness towards the needs and requirements that characterize their field of professional activity, a lucid perception of their preferences and expectations regarding the implementation of visualization techniques into their working environment, as well as an unambiguous opinion, based on their experience and professional expertise, concerning the contributions that virtual heritage applications could offer to their areas of activity. If such is the case, we expect most of the questionnaires returned by subjects belonging to the same target group to exhibit similar answering patterns. Such results would indicate that the sampled groups do feature a sufficient internal coherence to be considered as a representative indication of the tendencies that characterize their actor s classes, hence encouraging the generalization of the information that can be extracted from each of those groups.

135 134 Methodological approaches for the protection of cultural heritage in the digital age 4.5 Hypotheses Proceeding from the definition of our problem statement, an important step that needs to be taken consists is the formulation of a few hypotheses that are connected to the research questions previously identified. As it is intuitively clear, the primary working hypothesis that serves as a base for the definition of our set of research questions concerns the link between the preferences expressed by the participants to the experimental assessment and the group to which they belong. Since each of the selected groups is linked to a particular dimension that characterizes the protection of cultural heritage, it is natural to surmise that depending on their field of activity, each of those groups also exhibits different priorities while pursuing their own goals. We therefore infer the following primary hypotheses in relation to the first four stated research questions: o The priorities expressed by the subjects of the study are specific to the actor s class to which they belong. o The needs and expectations of each end user group, in respect to a visualization strategy that targets a heritage item, depend on the priorities that are inherent to the dimension of the protection of cultural o heritage to which they are connected. Each of the considered end users groups perceives the usefulness of a given visualization approach in relation to its degree of adaptation to their expressed needs and expectations. o The preferences expressed by the subjects targeted by our analysis, and the elements that constitute a successful virtual cultural heritage application from their point of view, are directly related to the actor s class to which they belong. Proceeding from these general hypotheses, we can define some secondary hypotheses which are related to the remaining research questions: o Each end user group should display some sort of preference towards one or more specific visualization approaches which they consider the most likely to satisfy their preferences. o Since the aspects that each group considers essential in order to constitute a successful virtual cultural heritage application might differ, given that their priorities are dissimilar, it is likely that their preferences will be expressed through the identification of elements pertaining either to: a cluster of necessary core requirements a group of nice to have but ultimately optional traits a set of unnecessary characteristics toward which they are indifferent Hence, while the core requirements might be dissimilar from group to group, it is nonetheless foreseeable, whenever the optional traits are concurrently considered, that some overlap might exist between the preferences that are expressed by each of the considered groups.

136 Methods 135 o By considering such overlaps it should be possible to determine an optimum balance, and therefore one or more visualization strategies, that could simultaneously satisfy the key requirements expressed by several of the considered groups. In order to determine such optimal approaches, we plan to integrate the preferences that are expressed by those groups as a new layer in the conceptual models that have been introduced in section 3.4. The definition of such layer should then allow the visual identification of intersecting regions that should contain the visualization strategies that satisfy this balance. These hypotheses, which are defined on the basis of the theoretical framework that has been developed in the previously presented chapters, will have to be empirically verified in the frame of our survey. Although some of these hypotheses might not be confirmed nor supported by our study, it is nonetheless important to note that their primary usefulness is not to anticipate without error the final outcome of the experimental assessment, but to serve as a conceptual tool instead. Their role is to provide a theoretical prediction of the results of the surveys and to serve as a frame of reference for the interpretation of the experimental data though the analysis of the deviations that may appear between the theoretically expected outcome and the empirically observed behavior of each of the selected groups. 4.6 Questionnaires In this section, the design choices that were applied to the creation of the tools that were used for the data gathering process, more specifically, the details pertaining to the steps that were taken during the definition of the questionnaire, available as appendix 2, will be overviewed. As such, the dimensions, indicators and variables that were defined to this end will be introduced. Finally, alongside the identification of the tools that are likely to be employed in the frame of the analysis that will be presented in chapter 5, as well as the results we expect to obtain, will be defined Dimensions The first step in the creation of a questionnaire is the hierarchical definition of the elements that pertain to the survey [D ASTOUS02]. To assist such effort, we can deconstruct the more abstract levels of our questionnaire in dimension and subdimensions. The identified dimensions, introduced by table 4, constitute the general areas that are targeted by our assessment and need to be examined to provide a more detailed vision of the studied object. It is interesting to note that most of the sub-dimensions that were employed to measure the personal preferences expressed by the interviewed subjects in relation to the implementation of virtual heritage applications do coincide with the continua that were identified during the definition of the taxonomical models that were presented in chapter 3.

137 136 Methodological approaches for the protection of cultural heritage in the digital age Dimensions Sub dimensions Sections of the Questionnaire Socio demographic information related to the interviewed subject Demographic information Education Technological affinity Section A Section B Section H Personal preferences concerning the use of visualization technologies targeting cultural heritage items Characteristics of the target group to which the interviewed subject belongs. Virtuality Interactivity Visual consistency Precision Automatism Usefulness Domain of activity Practical needs and requirements Degree of penetration of new technologies Section C and Section D Section E Section F Section G Table 4: dimensions and sub-dimensions employed for the creation of the questionnaire By measuring the degrees of virtuality, interactivity, precision, visual consistency and automatism that the participants to the questionnaire would prefer in a virtual heritage application, it should be possible to integrate such information in our conceptual framework, as introduced in section 4.5, therefore offering the chance to visually determine any eventual overlaps and common tendencies that would highlight the visualization approaches that simultaneously satisfy the needs and requirements of multiple actors Indicators While the dimensions of a questionnaire represent the general concepts and areas of interest that are targeted by the survey, the indicators serve as a tool to measure the affinity of the interviewed subjects towards a particular dimension. Since each dimension is measured by several indicators, as presented in table 5 and table 6, the analysis of the answering pattern of the participants to the survey will help in establishing their profile in relation to the previously indentified research questions.

138 Methods 137 Sub dimensions Indicators Questions Demographic information Age Gender Q01 Q02 Education More advanced degree obtained Q03 Technological affinity Domain of activity Practical needs and requirements Degree of penetration of new technologies Familiarity with technical terms use of information and communication technologies Affinity with real time graphics software Affinity with computer graphics hardware Professional field of activity Dimension of the protection of cultural heritage linked to the professional activity Degree of user friendliness required for a new professional tool Features that a professional tool must provide Type of data employed in the frame of the professional activity User friendliness vs. financial cost vs. quality of the results Frequency of use of new technologies for professional tasks Tasks for which new technologies are employed Age of the computers employed for professional use Type of Operating System used awareness towards the professional use of new technologies Q64 Q65 Q66 Q67 Q52 Q53 Q54 Q55 Q56 Q58 Q59 Q60 Q61 Q62 Q63 Table 5: indicators characterizing the professional field and socio-demographic information related to the interviewed subjects As such, in order to further deconstruct our questionnaire and introduce a more concrete representation of the concepts it seeks to analyze, each indicator needs to be transposed in the form of a specific question. The objective of each question is to measure the response of the interviewed subjects in respect to a specific variable that contributes to a particular dimension.

139 138 Methodological approaches for the protection of cultural heritage in the digital age Sub dimensions Indicators Questions Virtuality Interactivity Visual consistency Precision Usefulness of a 2D visualization techniques Q05 Usefulness of a 3D visualization techniques Q06 Usefulness of anastylosis Q07 Historical context of the visualization Q08 Usefulness of a visualization targeting the past Q09 Usefulness of a visualization targeting the present Q10 Importance of the Visualization of modern elements Q27 Preference between manual and CAD methods Q40 Preference between 2D and 3D visualizations Q41 Preference regarding the mixed use of real and virtual elements Q48, Q49 Importance to include the possibility to access external data Q50 Degree of interactivity featured by virtual humans Q31 Interactivity vs. precision vs. beauty Q33 Importance of Interactive visualization Q34 Type of preferred use of a virtual heritage application Q37 Preferred degree of Interactivity in virtual heritage application Q38 Preference concerning static and interactive models Q43 Importance of visual beauty Q12 Beauty vs. precision Q15 Importance of the realistic representation of materials Q21, Q22 Importance of the introduction of realistic lighting Q23, Q24 Importance of the visualization of the geographic context Q25, Q26 Importance of applying virtual restoration to damaged surfaces Q29 Preference regarding the geographic environment Q42 Preference regarding the inclusion of virtual humans Q45 Preference regarding simplified and complex illumination Q46 Preference regarding de the degree of visual realism Q47 Precision vs. scientific validity vs. functionality vs. beauty Q11 Importance of precision Q13 Importance of scientific of historical validity Q14 Extent of the virtual additions Q16, Q19 rigor necessary to implement visualization hypotheses Q17 Importance of the inclusion of missing elements Q18 Importance of the inclusion of small elements Q20 Importance to identify additions based on hypotheses Q28 Importance to introduce virtual humans Q30, Q32 Preference concerning detailed and simplified models Q44 Preference regarding the inclusion of sound Q51

140 Methods 139 Automatism Usefulness Maximum time willing to invest or wait to obtain a virtual heritage application fully meeting the necessary requirements Quality of the results vs. speed of to produce them Areas benefitting from virtual heritage applications Type of uses that are best suited for virtual heritage applications Q39 Q57 Q35 Q36 Table 6: indicators characterizing the preferences of the interviewed subjects concerning the use of visualization technologies in the frame of virtual heritage applications The indicators that describe the end user preferences were determined by employing as a base the main axes highlighted by table 7. While table 6 presents all the indicators that were employed to measure such preferences, it is interesting to note that section D of the questionnaire was specifically designed to assess the opinions of the interviewed subjects concerning those particular axes. The exploration of the information that is contained in such section of the questionnaire should then provide a valid starting point for our analysis. Sub dimensions Main indicator axes Questions Virtuality Virtual Reality Augmented Reality simulations 2D visualizations 3D visualizations Computer Generated Hand drawn restitutions Presentation of visual information only possibility to access external data Real time non real time simulations Q48 Q41 Q40 Q50 Q43 Interactivity Scripted exploration free form exploration of the environments Q34* Scenario based exploration non scenario based exploration Q31* Visual consistency Inhabited virtual environments uninhabited virtual environments Photo realistic non photo realistic representations Q45 Q47 Precise approximate restitutions Q44 Precision Geographically contextualized geographically not contextualized restitutions Q42 Realistic lighting simulation basic illumination models Q46 Automatism Rapid prototyping long development cycles Q39* Q57* Usefulness Didactic ludic presentation of the content of the restitution Q36* Table 7: main axes that have been employed for the selection of the indicators. The * designates the indicators that are not explicitly measured by section D of the questionnaire

141 140 Methodological approaches for the protection of cultural heritage in the digital age Finally, by referring to the third column of table 6, it is important to note that some indicators are measured by more than one variable. Such redundancy has been introduced to verify the coherence of the opinions expressed by the subjects targeted by the survey. Since the main object of these questions is similar, although slightly differently formulated and measured, the direction and intensity featured by their associated variables should be comparable Independent variables Another important step in the construction of a survey tool is to clearly identify the dependent and independent variables that will be considered for the subsequent analysis. The observation of how a variation of the independent variables, also referred as explanatory variables, impacts the dependent variables, also referred as explained variables, by means of eventual correlations is one of the central elements that can be highlighted by a statistical study. In the frame of our experimental assessment we have considered as independent variables the indicators that appear in table 8. Sections Indicators Type of variable Section A Age Gender Interval scale Nominal Section B More advanced degree obtained Ordinal Section H Section E Section F Section G Familiarity with technical terms use of information and communication technologies Affinity with real time graphics software Affinity with computer graphics hardware Professional field of activity Dimension of the protection of cultural heritage linked to the professional activity Degree of user friendliness required for a new tool Features that a professional tool must provide Type of data employed in the frame of the professional activity User friendliness vs. financial cost vs. quality of the results Frequency of use of new technologies for professional tasks Tasks for which new technologies are employed Age of the computers employed for professional use Type of Operating System used awareness towards the professional use of new technologies Multiple response Multiple response Ordinal Ordinal Nominal Nominal Ordinal Nominal Nominal Nominal Ordinal Multiple response Ordinal Ordinal Ordinal Table 8: indicators chosen as independent variables and their type

142 Methods 141 Since ordinal variables often cause confusion in data analysis, we chose to treat them, whenever necessary, as interval scale variables by assuming that their underlying scale is continuous and that the intervals are evenly spaced across such scale [UNESCO08b]. Likewise, in order to handle indicators that are characterized by multiple responses, the decision to represent them by means of dummy ordinal variables, and subsequently to treat them as interval scale variables, has also been preferred. It is important to mention that while the variables that appear in Section A and Section B of the questionnaire have mostly been included to provide a base for a descriptive study of the considered population sample, the main independent variables that characterize the survey are contained in Section E. Nevertheless, we do not exclude a priori that such variables could impact the preferences expressed by the interviewed subjects. Furthermore, in order to simplify the analytical study of the results, we will attempt to aggregate most of the ordinal indicators of section G and section H to build two continuous scale variables that respectively measure the technological affinity of the interviewed subjects and the degree of technological penetration that characterizes their working environment. Finally, it is essential to note that the indicators introduced in section F of the questionnaire will not directly serve as independent variables in the frame of our analysis. The nominal data provided by such section will be employed to characterize each of the considered actor s classes as well as to put into perspective the outcome of the surveys. With the evident exception of the general public, such information should allow a better assessment of the professional needs and requirements of the selected user groups regarding the optimal form that their preferred visualization approach should posses to permit its seamless integration in their professional working pipelines Age The selection of the age as independent interval scale variable, besides the descriptive information it can provide on the population sample, is motivated by the effect it could eventually have regarding the preferences expressed by the interviewed subjects. It is in fact possible that the more a subject is old, the more his positive inclination towards the use of virtual heritage applications in the frame of his professional activity could be diminished. Such behavior could originate from the fact that modern visualization technologies have only recently found their place in the field of cultural heritage protection: it is therefore plausible to assume that someone that has spent most of his professional career by working with traditional methods might express a resistance towards the adoption of such digital technologies Gender While the nominal variable gender has been introduced to mainly provide a more complete overview of the demographic profile of the population sample, we do not exclude that a correlation between such variable and the variations in the set of independent variables might be eventually be observed. If such is the case, we however believe that any correlation between the gender of the interviewed subjects and the preferences they express concerning virtual heritage

143 142 Methodological approaches for the protection of cultural heritage in the digital age applications is unlikely to be significant. In fact, a correlation indicating such a phenomenon might be biased by the fact that most of the professional categories that are targeted by our survey do not exhibit a balanced gender distribution Education While the ordinal variable education has been introduced to better characterize the population sample, it is however possible that it might exert a discernable effect on the independent variables. It is not excluded that a subject that benefits from a higher education could exhibit a more pronounced degree of openmindedness towards new technologies and the contributions that they can offer whenever applied to his domain of activity. Such inclination could therefore positively influence the preferences expressed by those subjects towards virtual heritage applications as well as their willingness to implement such techniques in their working environment Technological affinity The inclusion of technological affinity as independent continuous scale aggregate variable is based on the assumption that the subjects that already display a certain personal inclination towards digital tools are likely to positively consider virtual heritage applications. As such, it is plausible that such subjects, already accustomed by personal preference to the everyday uses of modern information and communication technologies, would be more receptive towards the introduction of digitally based visualization strategies than the subjects that are characterized by a limited exposure to such technologies. Although the effects of this variable could be considered similar to the ones eventually exerted by the variable age, it is important to note that they are in principle independent from one another: a given subject could exhibit a high degree of technological affinity independently from its age Domain of activity As previously introduced in section 4.5, the dimension of the protection of cultural heritage to which the activity of the interviewed subject is connected represents the main independent variable that is considered for this study. Since the main actors and their areas of intervention have already been identified while defining the target groups of the experimental assessment, this nominal variable has a known value that is not directly related to the answers given by the interviewed subject. However, while Q53 of section E of the questionnaire does not contribute per se to the definition of an independent variable, it could nevertheless provide an interesting perspective on the perception that the interviewed subjects have of the role that their professional activity occupies in the frame of the protection of cultural heritage. Moreover, Q52 of section E of the questionnaire has been introduced to offer a better insight on the selected population sample by providing some information on the professional profile of the interviewed subjects.

144 Methods Technological penetration of the working environment A further continuous scale aggregate variable that has been considered in the frame of this study is the technological penetration that characterizes the working environment of the interviewed subjects. Similarly to the variable that measures the technological affinity, it is conceivable that the subjects that are already accustomed in the frame of their professional activity to the frequent use of modern digital technologies could exhibit a positive inclination towards the implementation of virtual heritage applications Dependent variables The second necessary component that provides a clear structure to our study is the identification of a set of dependent variables, which are presented in table 9 and table 10. In the frame of our analysis, those variables, introduced by section C and section D of the questionnaire, as well as by one question of section F, specifically measure the preferences of the interviewed subjects regarding virtual heritage applications. By observing the values of the indicators that are linked to such variables, we will attempt to infer a preference profile for each of those subjects. The subsequent task will then consist in establishing whether or not a correlation exists between such answering patterns and the independent variables that were considered for this study. As such, the hypothetic causality link between the personal profiles of the interviewed subjects end their preferences regarding the use of digital visualization techniques will be explored. Consequently, it should be possible to highlight whether or not the variation of a given independent variable, such as for instance a different level of technological affinity, accounts for a significant modification in the measured preferences. For reasons similar to the ones previously mentioned regarding the independent ordinal variables, we decided to treat the dependent ordinal variables, whenever necessary, as interval scale values. However, a different option has been preferred concerning multiple choice variables. In this case, principally due to the nature of the concerned questions, such dependent variables have not been represented by means of dummy ordinal values, but they have been injected instead into dummy nominal values that represent the category that better describes the answering pattern that was given by the interviewed subject. Subdimensions Indicators of section C Type of variable Virtuality Usefulness of a 2D visualization techniques Usefulness of a 3D visualization techniques Usefulness of anastylosis Historical context of the visualization Usefulness of a visualization targeting the past Usefulness of a visualization targeting the present Importance of the Visualization of modern elements Ordinal Ordinal Ordinal Nominal Ordinal Ordinal Nominal

145 144 Methodological approaches for the protection of cultural heritage in the digital age Interactivity Visual consistency Precision Automatism Usefulness Degree of interactivity featured by virtual humans Interactivity vs. beauty Importance of Interactive visualization Type of preferred use of a virtual heritage application Preferred degree of Interactivity in virtual heritage application Importance of visual beauty Beauty vs. precision Importance of the realistic representation of materials Importance of the introduction of realistic lighting Importance of the visualization of the geographic context Importance of applying virtual restoration to damaged surfaces Precision vs. scientific validity vs. functionality vs. beauty Importance of precision Importance of scientific of historical validity Extent of the virtual additions rigor necessary to implement visualization hypotheses Importance of the inclusion of missing elements Importance of the inclusion of small elements Importance to identify additions based on hypotheses Importance to introduce virtual humans Maximum time willing to invest or wait to obtain a virtual heritage application fully meeting the necessary requirements Quality of the results vs. speed of to produce them Areas benefitting from virtual heritage applications Type of uses that are best suited for virtual heritage applications Ordinal Nominal Ordinal Multiple choice Multiple choice Ordinal Nominal Ordinal* Ordinal* Ordinal* Ordinal Nominal Ordinal Ordinal Ordinal Ordinal Ordinal Ordinal Ordinal Ordinal* Ordinal Nominal** Multiple choice Multiple choice Table 9: indicators of section C of the questionnaire chosen as dependent variables and their type The * designates the indicators that are measured by means of two variables The ** designates the indicators belonging to section F of the questionnaire In order to simplify the analysis of the gathered data, an aggregation of all the ordinal dependent variables identified in section C and section D of the questionnaire will be attempted. Such process should allow the construction of five distinct continuous scale variables: each of these variables represents a particular dimension appearing in the conceptual classification models that were introduced in chapter 3. Such aggregate variables will therefore represent the sub-dimensions of the questionnaire that characterize the preferences that were expressed by the interviewed subjects regarding virtual heritage applications.

146 Methods 145 Sub dimensions Indicators of section D Type of variable Preference between manual and CAD methods Ordinal Virtuality Preference between 2D and 3D visualizations Ordinal Preference regarding the mixed use of real and virtual elements Ordinal Importance to include the possibility to access external data Ordinal Interactivity Preference concerning static and interactive models Ordinal Preference regarding the geographic environment Ordinal Visual consistency Precision Preference regarding the inclusion of virtual humans Ordinal Preference regarding simplified and complex illumination Ordinal Preference regarding de the degree of visual realism Ordinal Preference concerning detailed and simplified models Ordinal Preference regarding the inclusion of sound Ordinal Table 10: indicators of section D of the questionnaire chosen as dependent variables and their type Finally, the separate analysis of the remaining dependent nominal variables should provide a further insight on specific aspects that the groups targeted by the experimental assessment consider important concerning the application of digital visualization techniques to a cultural property Exploitation of the data To exploit the data that has been gathered by means of the questionnaires, one of the first tasks we need to complete is to aggregate several indicators in order to construct fewer, and simpler to handle, continuous scale variables that facilitate our analysis. To assist such procedure, the constitution of a Likert scale [LIKERT32] to represent the degree of preference that is exhibited by the interviewed subjects towards each of the considered dimensions has been preferred. However, the questions introduced in our survey do not feature the five-level formatting that typically characterizes the Likert items. The implementation of a four-level formatting has been preferred instead by removing the neutral position, usually noted as neither agree nor disagree, to force the interviewed subject to take sides and express a clear position. Furthermore, other aggregation techniques, such as the Inglehart classification method [INGLEHART77] or the Katz [KATZ55] and Lazarsfeld [LAZARSFELD44] ranking strategy, have been discarded due to the high number of variables that were used in our study: such large quantity of variables would have created a combinatory complexity far too great to be effectively handled. Additionally, factor analysis, and more specifically principal component analysis, have been equally discarded due to the high number of observations they require as input to produce reliable results.

147 146 Methodological approaches for the protection of cultural heritage in the digital age Once the preliminary frequency and reliability analysis of the items that constitute such scales have been run, and after having weighted such items to produce comparable scales, we will conduct most of or study by means of crosstabulation and mean comparison techniques, as well as by running an analysis of variance whenever useful or necessary. In order to partially compensate for the small size of our sample, exact tests, such as for instance the Fisher exact test [FISHER70], might be employed wherever applicable. However, given the available number of observations, it is possible that their probability of error might exceed the 5% threshold usually retained in social sciences to determine the significance of such tests. Therefore, depending on the case, their relevance will have to be assessed while performing the interpretation of the results. Finally, by exploiting the appropriate graphical representations, tests and coefficients, and depending on the level of measure of the considered variables, we will attempt to empirically highlight the preference profiles of each of the groups that participated in our survey. Likewise, we will try to measure the influence that is eventually exerted by independent factors other than the actor class to which the interviewed subjects belongs, such as for instance their level of education or their age Expected results In this section a synthetic overview of the expected results of our survey will be presented. While the verification of our primary hypothesis, as stated in section 4.5, and further summarized by figure 32, should happen to be relatively straightforward, provided that the concerned variables can be easily aggregated, its validation is nonetheless a required step to determine the pertinence of our secondary hypotheses. Actor class Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Priorities Investigation + Expertise Defense + Conservation Dissemination + Documentation targets for the Dissemination technological mediators Specific needs and expectations Preference for a given visualization strategy Figure 32: relation between actors classes, their priorities, their needs and expectations, and their preferences concerning virtual heritage applications

148 Methods 147 Since we assume that the conceptual relations between a given actor class, its priorities, its needs and expectations and its preference concerning a given virtual heritage application all have transitive properties, in order to validate our primary hypothesis it is therefore only necessary to ascertain the significance of the relation between the actor class and its expressed preferences. Actors active in cultural heritage protection activities Preferred dimensions characterizing a virtual heritage application Necessary Optional unnecessary Content Holders Precision Visual consistency Field Personnel Interactivity Automatism Interactivity Automatism Precision Visual consistency Cultural Intermediaries Visual consistency Interactivity Automatism Precision Content Consumers Visual consistency Interactivity Precision Automatism Digital Intermediaries Precision Interactivity Visual consistency Automatism Table 11: expected results concerning the dimensions characterizing a virtual heritage application favored by each of the actors targeted by the experimental assessment Consequently, a valid profile of the preferences of each of the considered actors could be determined by establishing which dimensions that characterize a virtual heritage application they favor the most. The expected results concerning such preferences, which are related to the second secondary hypothesis introduced in section 4.5, are summarized by table 11 on the basis of the elements that were previously discussed in section 2.3 and in section 2.4. Visual Consistency and Precision Interactivity Automatism Figure 33: expected area of preference characterizing content holders

149 148 Methodological approaches for the protection of cultural heritage in the digital age Since such actors all feature different priorities, as introduced by figure 32, the dimensions that they consider necessary, optional or unnecessary should depend on the specific needs that are inherent to the role they play in the frame of the protection of cultural heritage. As such, content holders should favor geometrical precision in order to support their activities, which are connected to the investigation of cultural heritage and, in this particular case, an elevated degree of interactivity or automatism should prove unnecessary to assist their tasks. Field personnel, on the other hand, should exhibit a preference towards interactivity, since their area of intervention requires the ability to quickly access and display clear and pertinent spatial information concerning a given heritage site. Moreover, if we consider the activities that are related to the defense of cultural heritage, and more specifically the constitution of a security documentation for a specific site, a certain degree of automatism should also be welcomed to assist such task. In such case, a high precision or an elevated visual consistency should prove unnecessary to support their activities. Visual Consistency and Precision Interactivity Automatism Figure 34: expected area of preference characterizing field personnel Cultural intermediaries are instead expected to display a preference regarding high interactivity and visual consistency in order to appeal to the general public, which is the target of their disseminative activities.. Visual Consistency and Precision Interactivity Automatism Figure 35: expected area of preference characterizing cultural intermediaries

150 Methods 149 However, if we consider, for instance, a museum that is willing to create an online exposition of its collections, the possibility to develop and further update such virtual heritage application by investing a reasonable amount of time, and therefore implying an acceptable degree of automatism, could also be important. In either case, high geometrical precision should prove unnecessary to support an effective dissemination of the information that pertains to a given cultural heritage item. Content consumers, since visually pleasing images and interactive activities usually stimulate the attention of the viewer and facilitate the learning process, are likely to exhibit a positive inclination towards virtual heritage applications that exhibit high visual consistency and high interactivity. However, due to their high expectations for visual realism [Addison 2000], it is not unreasonable to anticipate that they will display an equal preference regarding the visualizations that feature a good degree of geometrical precision. Since content consumers constitute the target of the disseminative activities and since they do not actively participate in the creation of virtual heritage applications, the amount of time that is necessary to produce such content, and therefore the degree of automatism that characterizes such efforts, should not affect their preferences. Visual Consistency and Precision Interactivity Automatism Figure 36: expected area of preference characterizing both content consumers and digital intermediaries Digital intermediaries, since they are often in direct contact with recent technologies, and since they are in the ideal position to implement new approaches that could push the boundaries of the state of the art in virtual heritage, are equally expected to exhibit a preference concerning the applications that are characterized by high interactivity, high visual consistency and high precision. However, it should be noted that such consideration might not happen to be appropriate in cases where other aspects could be preferred instead, due for instance to specific time constraints or particular objectives that are to be met by the developed virtual heritage application. Nonetheless, whenever they are given the chance, we believe that digital intermediaries are likely to strive for the

151 150 Methodological approaches for the protection of cultural heritage in the digital age enhancement of all the dimensions that characterize a virtual heritage application at the expense of its degree of automatism. Such profiles should highlight the type of visualization strategy and content creation approach that is the most suited to match the needs and expectations of each of the groups that have been targeted by our experimental assessment, as they should highlight the technologies that are the more adapted to assist and support the activities carried out by such groups. The expected results concerning such preferred approaches, which are related to the first secondary hypothesis stated in section 4.5, are summarized by table 12 on the basis of the conceptual framework introduced in section 3.2 and in section 3.3, as well as on the basis of the elements that were introduced by table 11. Actors active in cultural heritage protection activities Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Preferred type of visualization strategy Renderings simulations Digital catalogs Digital panoramas Semantically supplemented 2D images Real time AR simulations Semantically supplemented 3D models Digitally augmented movies Renderings Videogames Stereoscopic visualizations Real time VR simulations Renderings Preferred type of content creation approach Spline surfaces 3D scanning Image based modeling Digital photography Polygonal modeling Image based modeling Polygonal modeling Spline surfaces Spline surfaces Polygonal modeling Table 12: expected results concerning the visualization strategies and content creation approaches preferred by each of the groups targeted by the experimental assessment By integrating the answering profile of each considered actor as a new layer into the taxonomical spaces introduced in chapter 3, and by considering the visualization strategies and content creation approaches that are favored by the each of the interviewed groups, it should be possible to empirically identify some overlaps that point to specific areas of the models which contain virtual heritage applications that could simultaneously satisfy the preferences expressed by several actors at once. The expected results concerning such areas of preference, which are related to the third secondary hypothesis stated in section 4.5, are summarized by figure 37 on the basis of the taxonomical models presented in section 3.2 and in

152 Methods 151 section 3.3, as well as on the basis of the elements highlighted by table 11 and by table 12. As highlighted by the overlapping areas identified in figure 37, real time VR simulations (item 10) and semantically supplemented 3D models (item 15a) should theoretically provide an acceptable trade-off regarding the needs and expectations expressed by all the considered actors. Similarly, polygonal modeling and image based modeling approaches should equally represent the most suited content creation techniques that would satisfy the practical requirements that characterize such actors in respect to those two visualization strategies. Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries A. Polygonal modeling B. 3D Scanning C. Image based modeling D. VRML modeling E. Spline surfaces F. Digital photography A Visual Consistency And Precision E a Renderings 8. Digital Catalogs 9. Digital panoramas 10. Real time VR simulations 11. Stereoscopic visualizations 12. Videogames 13. Real time AR simulations 14. Digitally augmented movies 15. Semantically supplemented 2D photographs 15a. Semantically supplemented 3D models B Interactivity D Figure 37: expected overlaps between the areas of preference featured by the actors studied in the frame of the experimental surveys 4.7 Interviews Automatism F In this section the design choices that were applied during the preparation and the conduction of the personal interviews, which were targeting selected individuals belonging to each of the studied groups, will be briefly overviewed. Furthermore, specific aspects concerning the choice of the interviewed subjects, the exploitation of the gathered information and the expected results will be presented. C

153 152 Methodological approaches for the protection of cultural heritage in the digital age Objective Since only a very limited number of subjects were selected as targets for the personal interviews, the general idea behind them was not to accumulate enough data to allow a rigorous analysis of their statements, but to gain instead a valuable insight on their stand regarding the place, role and usefulness of virtual heritage applications in the frame of their domain of activity. As such, these interviews were planned to serve as a complementary tool to support the interpretation of the data that was extracted from the questionnaires. Moreover, due to the relatively small population sample that was employed for the data gathering, such interviews should assist in better understanding the possible deviations that might eventually appear while processing the questionnaires Size of the sample As previously introduced in section 4.3.2, the size of the sample that was employed for the interviews is very limited. Despite the fact that only two subjects per group, as introduced by table 13, were selected to supplement our study with an expert opinion about the general priorities that characterize their own specific areas of professional activity, we do believe that such reduced number should nevertheless suit the objective of this part of the study. Actors active in cultural heritage protection activities Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Interviewed Subjects 1 university professor of archeology 1 university professor of ancient roman history 1 captain of the cultural heritage protection section of the Swiss civil protection 1 restoration specialist experienced in interventions targeting cultural heritage buildings 1 secondary school history teacher 1 director of a museum of history and archeology 2 people not belonging to other actors classes 1 university researcher specialized in informatics and computer graphics 1 computer graphic s artist having experience in virtual heritage projects Table 13: interviewed subjects and the actor s class to which they are linked In order to obtain a more comprehensive overview of the preferences, needs, priorities, doubts, expectations and requirements of each actor s class concerning the use of visualization technologies in the frame of cultural heritage

154 Methods 153 related activities, one person issued from each of the identified groups has been included in the selected sample Preparation and protocol of the interviews Each of the interviews has been conducted face to face over a period of time ranging from forty five to sixty minutes. While the general themes that were developed during the encounters were fixed and selected prior to the interviews, each particular discussion was planned as a free form exploration of the main subjects targeted by our study. Thus, the subjects participating in our assessment were not forced to follow a rigorous question-answer format while formulating their opinions. Nonetheless, some structure was provided by the fact that certain aspects of interest were introduced by general questions, following the same axes explored by the questionnaire, and addressed to the interviewed subjects regarding particular topics that are connected to their professional environment, their working pipeline and their views concerning the usefulness of virtual heritage applications. In order to maintain the focus of the interviews and keep them from expanding beyond the boundaries that define our study, further guiding questions were formulated on the fly using as base the statements provided by the interviewed subjects. Furthermore, special attention was given in order to avoid influencing the opinions expressed during the discussions. As such, caution has been exerted to conceal any hint regarding both the expected answers and the importance that specific elements might have in the frame of our study, as well as any unnecessary insistence on specific topics and themes of interest that were spontaneously introduced by the interviewed subjects Exploitation of the data The opinions gathered while conducting the personal interviews are not meant to be exploited in the frame of a structured investigation by means, for instance, of word frequency analysis. Nonetheless, such data should prove useful to guide and assist, whenever necessary, the interpretation of the information extracted from the questionnaires, as well as to elucidate unclear elements that might eventually appear while performing the analysis of such data. In order to allow a seamless integration of this information during the analysis of the questionnaires, the opinions offered by the interviewed subjects have been transposed in the form of key statements and bulleted lists which summarize their points of view regarding specific aspects that are related to the implementation of virtual heritage applications in their working pipelines. Since a face to face discussion is more likely to provide an insight into the motivations that underlie the answers that were provided by the interviewed subjects than a questionnaire, the exploitation of such information should allow a more coherent assessment of the real needs, requirements priorities and expectations that characterize the class of actors to which they belong.

155 154 Methodological approaches for the protection of cultural heritage in the digital age Expected results Since the interviewed subject should provide a representative point of view that is shared by the majority of the group to which they belong, their statements are expected, for the most part, to support and corroborate the outcome of the analysis of the questionnaires. However, we do not exclude the possibility that their opinions might eventually stray from the tendencies inferred from the study of such data. If this eventuality should arise, its source might be rooted in the fact that a face to face discussion could encourage the interviewed subject to consider more carefully their positions than a series of multiple choice answers appearing in a questionnaire. Alternatively, such divergence might be explained by the fact that the interviewing process has increased their awareness regarding the usefulness of the proposed approaches. In either case, the gathered information should become a valuable asset for the explanation of the different answering patterns that could emerge while conducting the analysis of our questionnaires. Therefore, the outcome of the personal interviews will be employed as a conceptual key to allow a more coherent evaluation and interpretation of the preferences that were expressed by the user groups that participated in our experimental assessment. 4.8 Previous work It is important to note that the approach that has been chosen to assess and highlight the user preferences we aim to study finds its original roots and inspiration into the statistical analysis that was conducted in the frame of the INCO-MED FP5 European Projects ERATO (ICA3-CT ) and CAHRISMA (ICA3-CT ). However, while these projects have explored similar research directions, the methods used in those instances to construct and analyze their questionnaires were confined to the study of the audio visual components of a 3D restitution. Moreover, due to the even more limited size of the sample that was employed in the frame of those projects and due to the nature of the simple descriptive univariate statistics that were produced, the outcome of those studies does not lend itself to a broader generalization as we aim to offer. Starting from those same ideas, we thus improve such approach by proposing a broader, more robust and multidimensional survey strategy, as well as by introducing the use of bivariate analysis to better capture the complexity of the question at hand and allow for a seamless integration of the statistical results with our original taxonomical model. 4.9 Summary In this chapter, the methods that have been selected for the experimental assessment of the preferences expressed by the main actors connected to the protection of cultural heritage have been overviewed. As such, the main objectives and the research questions that motivate our work have been introduced alongside the hypotheses and the conceptual boundaries that define both our study and our survey.

156 Methods 155 Secondly, the specific groups that constitute the intended target of our assessment have been highlighted concurrently to the analytical tools that were chosen for the collection of the necessary experimental data. More specifically, the modalities concerning the conduction of the data gathering process, by means of a self administered questionnaire and a series of face to face interviews, have been introduced. Subsequently, the main assumptions upon which the further analysis of such data will be based have been presented. Furthermore, the main design choices that underlie the creation of the questionnaire, as well as the key aspects that concern the protocol of the personal interviews, have been detailed. Consequently, the dimensions, subdimensions and indicators that have been used as a base to create such tools have been discussed in conjunction with the definition of the dependent and independent variables that define the conceptual frame for the analysis of the collected data. Finally, the tools that were specifically selected for the study of the gathered data have been shortly overviewed and accompanied by the presentation of the expected outcome of our survey.

157 156 Methodological approaches for the protection of cultural heritage in the digital age

158 Analysis 157 Chapter 5. Analysis While chapter 4 has introduced the general methodological design choices, the statistical tools as well as the theoretical outcome we expect to find by studying our sample population in respect to the outlined research questions, this chapter will present the different steps that have been taken to transform, manipulate and examine the collected data alongside the empirical results that were obtained in the frame of the analysis of both the questionnaires and the interviews that were employed to determine the preference profile for each of the groups that participated in our survey. As a first step, the preliminary manipulations that have been applied to the raw data to facilitate its practical exploitation will be introduced, such as the recoding and weighting of the variables and their aggregation into continuous scales. Subsequently, a descriptive overview of each of the aggregate variables that have been constructed will be presented followed by a correlation analysis which should highlight the contribution of the selected independent variables to the preference profiles of the studied groups. Then, by retaining the pertinent variables, the final preference profiles for each of the considered groups will be constructed and plotted as a histogram on in order to illustrate the main differences and similarities between the elements that each of the studied actors considers as the most critically important for a useful visualization of a heritage item. Finally, the convergence, or divergence, between the empirically constructed preferences, as their eventual overlaps, alongside the theoretically predicted ones will be assessed while presenting some concluding remarks on the results of our study. 5.1 Variable recoding and weighting Since the questionnaire designed for our survey includes multiple choice answers, our first priority is to ensure that each recorded answer is correctly translated as a contributing factor to one of the main dependent or independent aggregate variables that have been employed in the frame of our analysis. As such, the variable values that are assigned to each of the answers, and the overall weight that each question has regarding its specific contribution to the aggregate scale it is associated with, should be addressed. While it has been preferred to consider the overall weight of all the variables as equal to 1, concerning their respective contribution to their associated aggregate scale, some recoding has nonetheless been necessary to standardize the values of some of the answers that were provided by the users target by our survey. This necessity mainly comes from the fact that not all the questions offer the same amount of possible choices to code their answers.

159 158 Methodological approaches for the protection of cultural heritage in the digital age Possible choices Raw value Recoded value Recoded Label Contributes to I do not know No (favor precision) Yes (favor interactivity) Not important at all Marginally important Rather important Very important Static images Simple models Realistic models No, never Rarely Sometimes Often Very frequently Missing Very low Very high Very low Low High Very high Very low Average Very high Very low Low Average High Very high (Precision +) Interactivity + Virtuality + Precision + Technological penetration + Table 14: example of variable value recoding As example of the value recoding that has been applied to the raw data we can refer to table 14: this table details some of the actions that have been carried out in order to transform and normalize the values associated with some sample questions that were proposing a different amount of possible answers. It is important to note that not all the questions are positively associated to exclusively one scale: in some rare instances some values have been used as a contributing factor to more than one scale. Those cases are generally the ones that are presenting only two choices and which are used to infer a particular preference for one dimension over another: the first row of table 14 shows that the same question is contributing both to the precision and to the interactivity scales. However, the values employed for the aggregation are not the same as it is presented by table 15: the value that is associated with the answer that was explicitly given by the participant to the survey is added to one scale while the complementary implicit value of the alternative choice is added to the other scale, Possible choices Recorded choice Contributes to Contributes to I do not know No (favor precision) YES Precision + 0 Interactivity Yes (favor interactivity) I do not know No (favor precision) Yes (favor interactivity) NO Precision Interactivity + 0 Table 15: example of variable contributing explicitly to one scale and implicitly to another one

160 Analysis 159 A second element that is important to stress is that most questions present the possibility to have missing values. In the cases where the questionnaire was returned by a participant in an incomplete state, i.e. presenting missing answers, the corresponding variables were not considered as contributing to the construction of the aggregate scales. 5.2 Data aggregation and reliability analysis After having normalized all the variable values that were associated with each of the answers coded in the questionnaire, in order to make them comparable with each other, and after having assigned a weight to each of the questions that were submitted to the sampled population during our survey, such data has been aggregated to construct a smaller and more manageable number of variables to facilitate its analysis. Subsequent to a preliminary study of the recorded data, we have chosen to retain two groups of variables to assess the user preference profiles: the first group consists of five dependent variables while the second one consists of five independent variables, as presented by table 16. Independent variables Dependent variables Domain of activity Virtuality * Age Interactivity * Gender Visual consistency * Education Precision * Technological affinity * Automatism * Table 16: variables retained in the frame of our analysis; * denotes a continuous aggregate scale variable In order to aggregate the data, an additive approach based on the Likert scale construction strategy [LIKERT32] has been employed by following the formula presented in figure 38. As a part of the data aggregation process, a reliability analysis, which is detailed in the remainder of this section, has been conducted on the items that were employed for the constitution of each of the considered scales. To assess the reliability of each scale, as well as the specific contribution of each of the variables that were assigned to each set, we have iteratively employed as main indicator the Cronbach's alpha value. As such, several variables have been progressively discarded from the initially selected set whenever their removal could have increase the alpha value of the scale they had been assigned to.

161 160 Methodological approaches for the protection of cultural heritage in the digital age i Q W Figure 38: top, aggregation formula used to construct the scale variables, where V= scale value, n=number of valid questions in the set, i=position of the question in the set, Q=variable value, W=weight of the question; center and bottom, example on a test dataset As a side note, it is important to note that since all the questions in our data set have been assigned values that range from 0 to 100 during the recoding phase, the final aggregate value for the additive scales is also a number that is included in such range. Such scores will be employed in section 5.5 as percentages which indicate the degree of preference that is expressed by each of the surveyed groups Technological affinity The first aggregate variable that has been constructed is the one that measures the user familiarity with modern digital technologies. This scale will be employed as an independent continuous scale variable, alongside age, gender, education and domain of activity, to verify if it exerts some kind of influence on the user preferences regarding the dimensions that characterize a digitally based visual restitution of a cultural heritage item: i.e. virtuality, interactivity, visual consistency, precision and automatism. The initial reliability analysis, presented in table 17, has been run on a selected group of nine questions, which were designed to offer a set of indicators to measure the degree of technological affinity that is exhibited by each of the participants to the survey. While the alpha value returned by the initial reliability test could already be deemed satisfying, we nonetheless have to note that it could be possible to improve it by discarding Q62, which presents the lowest item correlation score compared to the other variables in the set.

162 Analysis 161 Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N Items of Corrected Item- Total Correlation Squared Multiple Correlation Q Q Q Q Q Q Q Q Q Cronbach's Alpha if Item Deleted Table 17: technological affinity, initial reliability analysis This means that Q62 does not contribute significantly to the determination of a meaningful score if used to construct the technological affinity scale: it therefore needs to be removed from the initial set and a new reliability analysis should be run on the remaining items to see if a further improvement of such score is possible. Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N Items of Corrected Item- Total Correlation Squared Multiple Correlation Q Q Q Q Q Cronbach's Alpha if Item Deleted Table 18: technological affinity, final reliability analysis after deletion of the elements permitting the increase of the alpha value

163 162 Methodological approaches for the protection of cultural heritage in the digital age After a few iterations no more benefit would have been gained by discarding further items from the list: as such, a final set of variables has been determined as the base to construct the technological affinity scale. Each if the items that were included in the final list, presented in table 18, cannot be removed without causing a drop of the alpha value Interactivity The second continuous scale variable that has been constructed targets the level of preference that is expressed by the survey participants towards the degree of interactivity featured by a digitally based visual restitution in the frame of a cultural virtual heritage project. The interactivity scale will be treated as dependent variable, since we aim at determining if it is influenced by any of the selected independent variables. Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items Corrected Item- Total Correlation Squared Multiple Correlation Q Q33Int Q Q38a Q38b Q Cronbach's Alpha if Item Deleted Table 19: interactivity, initial and final reliability analysis After a preliminary reliability analysis that has been run on the initially selected variables, each representing an indicator that is associated with one question of the survey, as presented in table 19, all variables have been retained for the construction of the final scale, since no improvement of the alpha value would have been possible by discarding of any of them Virtuality Another continuous dependent scale variable that has been considered is the one that expresses the user preference regarding the degree of virtuality that is exhibited by a virtual cultural heritage visualization. While the analysis of this scale variable presents an overall mild interest in the frame of our study, it has nonetheless been included since it can highlight whether or not, and to which extent, the users favor an approach that is based on fully virtual visualization strategies, on manual approaches or on mixed technologies.

164 Analysis 163 Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items Corrected Item-Total Correlatio n Squared Multiple Correlatio n Cronbach' s Alpha if Item Deleted Q Q Q Q Q Q Q Q29Vi r Q Q Q Q Q Q Q Corrected Item-Total Correlatio n Squared Multiple Correlatio n Cronbach' s Alpha if Item Deleted Q Q Q Q Q29Vi r Q Q Q Q Q Q Q Table 20: virtuality, initial reliability analysis Table 21: final reliability analysis After having iteratively run the reliability tests on an initial set of fifteen items, as shown in table 20 and table 21, twelve variables were retained for the construction of the interactivity scale that will be used to further analyze the gathered data Visual consistency and Precision A further dimension which is targeted by the data aggregation process is the one that measures the user preference towards visualization approaches that feature a high degree of visual consistency or precision. Since these two elements can be regarded as two aspects of the same dimension, as far as we consider their respective roles and their influence on the four dimensional taxonomical space that was introduced in section 3, and while a continuous dependent scale variable has been constructed to verify their combined effect, they also have been measured separately as two distinct continuous scale variables to allow a

165 164 Methodological approaches for the protection of cultural heritage in the digital age better insight on which facet has more impact on the preferences expressed by the groups targeted by our survey. Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N Items of Corrected Item- Total Correlation Squared Multiple Correlation Q Q Q Q Q Q Q Q Q33Vis Q Q Q Q Q Q Q Q Q Q Q Q Q Q29Vis Q Q Q58P Cronbach's Alpha if Item Deleted Table 22: visual consistency and precision, initial and final reliability analysis Table 22 shows the initial set of indicators that were selected as variables in the construction of the scale that is aimed at measuring the combined effect of both

166 Analysis 165 precision and visual consistency; however, since this set cannot be reduced further without decreasing the alpha value, a total of twenty six items has been retained for the data aggregation process. After having built this first scale, both aspects that concurrently participate in its constitution have been tested in order to create two separate continuous dependent scale variables: the first measures the user preference regarding the degree of precision that is exhibited by a visualization attempt, as shown in table 23 and table 24, while the second measures their preference concerning its degree of visual consistency, as shown in table 25. Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items Corrected Item-Total Correlatio n Squared Multiple Correlatio n Cronbach' s Alpha if Item Deleted Q Q Q Q Q Q Q Q Q Q Q29 Vis Q Q Q58 P Corrected Item-Total Correlatio n Squared Multiple Correlatio n Cronbach' s Alpha if Item Deleted Q Q Q Q Q Q Q Q Q Q29 Vis Q Q Q58 P Table 23: precision, initial reliability analysis Table 24: precision, final reliability analysis

167 166 Methodological approaches for the protection of cultural heritage in the digital age Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N Items of Corrected Item- Total Correlation Squared Multiple Correlation Q Q Q Q Q Q Q Q Q33Vis Q Q Q Q Cronbach's Alpha if Item Deleted Table 25: visual consistency, initial and final reliability analysis Automatism The last continuous scale dependent variable that will be employed in the frame of our study is the one representing the preference of the users regarding the degree of automatism that is exhibited by a virtual heritage visualization approach. However, after having run some reliability analysis on the initially considered set of variables, some problems emerged from the fact that the chosen items where presenting a negative convergence. This is likely due to some inconsistencies in the answer patterns that were given by the surveyed subjects, as well as to some variable coding issues regarding some of the questions. To fix the problem, some items where recoded to display values that were progressing in the same direction as the other items in the set, while some indicators where completely discarded from the initially selected set of variables. The final recoded item list, shown in table 27, presents a limited number of variables and an alpha value that is quite low, thus indicating a less than ideal reliability for the scale variable that results from their aggregation. However, as the available data does not offer a viable alternative to construct a more reliable aggregate variable, since in order to build a more solid automatism scale a new survey would be necessary with the inclusion of questions more carefully designed to specifically measure such dimension, the resulting scale has therefore been retained in the frame of our analysis.

168 Analysis 167 Cronbach's Alpha Q58A R Cronbach's Alpha Based on Standardized Items N of Items Corrected Item-Total Correlatio n Squared Multiple Correlatio n Cronbach' s Alpha if Item Deleted Q57R Q54R Cronbach's Alpha Q58A R Cronbach's Alpha Based on Standardized Items N of Items Corrected Item-Total Correlatio n Squared Multiple Correlatio n Alpha if Item Delete d Q57R Table 26: automatism, initial reliability analysis Table 27: automatism, final reliability analysis As a side note it is interesting to note that the question with the highest inter item correlation, which also presents the most clear variation in the answers provided by each of the groups targeted by the survey, is the one that deals with their preference concerning the tradeoff between monetary cost and quality of the results. Table 28 shows that cultural intermediaries are the ones who are the more affected by such variable: they will therefore be the ones that are the more likely to prefer a visualization approach that offers a cheaper and a faster alternative to the strategies that involve an expensive equipment or long development cycles. Q58A_Recoded expensive, perfect results cheap,average results Field Personnel 83.3% 16.7% Cultural Intermediaries 37.5% 62.5% Digital Intermediaries 78.6% 21.4% Total 72.5% 27.5% Table 28: monetary cost vs. quality of the results by actor group Ultimately, the pertinence of this scale variable will be tested in section 5.4, where we will verify if the variation of some of the independent variables produces a noticeable effect on its distribution, such as, for instance, whether or not different actor groups show a significantly different score regarding their preferred degree of automatism. Furthermore, to ensure a better reliability of the

169 168 Methodological approaches for the protection of cultural heritage in the digital age final preference profiles, these values will be weighed by employing the scores that derived from the interviews. 5.3 Univariate analysis Before employing the previously constructed scales in the frame of a bivariate analysis, it is interesting to take a quick look on each of the variables that were considered in our study by using a univariate descriptive approach in order to provide an insight into the structure of our sample population and their initial global preferences. The tables corresponding to each of the histograms that are presented in this sub-chapter are available in the descriptive analysis section of appendix Gender The first considered variable is the gender of the participants to our study as presented by table 29. Looking at the general distribution of the results we can notice that the sample population presents a clear imbalance in the representation of both genders. This outcome is due to the underrepresentation of women in some of the considered actor classes, such as is the case for digital intermediaries and field personnel. As such, due to the presence of skewed results, this particular variable will most likely prove inadequate in the frame of the bivariate analysis, since establishing any correlation with the dependent variables would probably be meaningless under such circumstances. Frequency Percent Valid Percent Cumulative Percent Valid Male Female Total Table 29: distribution of the gender variable in the sampled population Age The second variable that was retained is the age the participants to the survey. As illustrated by figure 39, there is a spike in correspondence to the population that is aged between 25 and 35 years. This is in part related to the specific demographic makeup of some of the groups that were targeted by our survey, such as the field personnel, and in part linked to the specific sample we had access to during the preparation of our study.

170 Analysis 169 Frequency Percent Valid Percent Valid between 21 and 25 years between 26 and 30 years between 31 and 35 years between 36 and 40 years between 41 and 45 years between 46 and 50 years between 51 and 55 years between 56 and 60 years Cumulative Percent More than 61 years Total Figure 39: distribution of the age variable in the sampled population Nonetheless, a broad range of values for this variable was preserved in the final sample: it would have been preferable, however, to obtain a curve whose apex was located between the 36 and 45 year old category Education The third variable that has been assessed is the education level of the people that undertook the self administered questionnaire. Figure 40 presents the highest academic degree held by those participants as well as their specific field of professional specialization. While the first distribution present itself as a bell curve, it is interesting to note that the second distribution features a higher concentration of answers noted as other. This is largely due to the fact that most of the people issued from the content consumer and the field personnel groups that have been interviewed do not have a specific training in any of the

171 170 Methodological approaches for the protection of cultural heritage in the digital age fields linked to the protection of cultural heritage that were proposed, such as archeology, architecture or restoration. Frequency Percent Valid Percent Cumulative Percent Valid Primary School Professional Maturity Secondary School Secondary Technical School Professional University University Bachelor Master Degree PhD Other Total Frequency Percent Valid Percent Valid History Archeology Architecture Artistical related Activities Restoration or Conservation Cumulative Percent Engineering Informatics Music Other Total

172 Analysis 171 Figure 40: distribution of the education variable in the sampled population Domain of activity The fourth variable that has been analyzed is the main independent variable around which our empirical study is constructed, namely the domain of activity or the actor class. As showed in table 30, the frequency distribution between the selected classes is mostly even by design. Frequency Percent Valid Percent Cumulative Percent Valid Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Total Table 30: distribution of the domain of activity variable in the sampled population However, it can be noted that the content consumer class constitutes a smaller percentage of the sampled population than originally intended. This slight underrepresentation is caused by the fact that a certain number of questionnaires distributed to this group have never been returned or have been excluded from the data input process due to their mostly incomplete state Technological affinity The fifth variable that was extracted from the gathered data is represented by the technological affinity scale. Figure 41 illustrates the distribution of the values of such variable in the sample population. While it is clear that the result is not a

173 172 Methodological approaches for the protection of cultural heritage in the digital age bell shaped curve, it is however interesting to note that the plotted frequencies suggest that this variable could have a polarizing effect. Figure 41: distribution of the education variable in the sampled population This would indicate that either someone is characterized by a low technological affinity, thus having little familiarity with new technologies, or is conversely characterized by a high degree of affinity, thus leaving a minority to occupy the mid zone that corresponds to an average degree of affinity towards modern technologies Frequency distribution of the aggregate scales The last variables of interest that need to be studied in the frame of our univariate analysis are the different dependent continuous scales that have been defined in section 5.2. Those variables, as depicted by figure 42, mostly feature a normal distribution with a slight skew on the right side of the histogram. This small anomaly is explained by the natural desire of each interviewed person to prefer, given the choice, a solution that maximizes the dimensions they deem the most important for their work, even when such extremes might not be required. This happens more often in cases where the inherent trade-offs that could be linked to their choices are not well understood. In other words, if someone requires a model featuring an average degree of precision, but he is given the choice to obtain one with an extreme amount of it, with no discernable trade-offs he could recognize, he would probably choose the second option despite the fact that it would exceed by far his basic requirements.

174 Analysis 173 Figure 42: distribution of the frequencies of the considered continuous dependent aggregate scale variables A special note has to be made about the histogram in the lower right of figure 42 that illustrates the frequency distribution of the automatism scale. Although a bell shape is discernable, with the presence of a slight skew towards the left, which would indicate that this dimension is not deemed by most of the considered groups as important as we had previously theorized, the overall uniform distribution could suggest that the sampled population does not exhibit any strong feelings towards neither ends of the spectrum that is defined by such scale.

175 174 Methodological approaches for the protection of cultural heritage in the digital age In this case, it would be possible to surmise that most of the user groups that were included in our study, exception made for the cultural intermediaries, are displaying a general indifference towards the automatism scale and its trade-off mechanism in relation to the other dimensions. Those actors would then be likely to consider automatism as an unnecessary factor in the definition of their preferences, and therefore turn their attention to other dimensions that they deem more important for their respective fields of activity. 5.4 Bivariate analysis We have identified five different independent variables that could eventually explain any observed variations in the user preference profiles codified by means of the dependent continuous variables that were introduced in section 5.2. These independent variables are the gender, the level of education the age, the technological affinity and the domain of activity. Since the results of the survey and the answers provided by the sample population were not uniform, our main hypothesis, which would explain such variation, is that each actor class exhibits specific preferences depending on the needs and expectation that are related to their specific field of activity in the frame of the protection of cultural heritage. However, we cannot discard the possibility that other factors could inform or influence those preferences. For this reason, in this section, we will analyze each of the selected independent variables and check if it is possible to identify any correlation between their variation and the one observed in the dependant variables. If these tests prove conclusive, we should be able to construct a preference profile by employing the pertinent variables that will be retained after this step Gender The first independent variable tested is the one that codifies gender. As surmised in section 5.3.1, and as shown by table 31, the p-value for Levene s test, as reported under the Sig. column, suggests that gender does not have an impact on the preferences expressed by the sample population that participated in our study.

176 Analysis 175 VIRTUALITY_SCALE INTERACTIVITY_SCALE PRECISION_SCALE VISUAL_CONSISTENCY_SC ALE PRECISION_VISUAL_AGGRE GATE_SCALE AUTOMATISM_SCALE Between Groups Within Groups Sum of Squares df Mean Square F Sig Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Table 31: ANOVA table analyzing the influence of the gender variable on the actor preferences Age The second independent variable analyzed is the one representing the age of the participants to the survey. As it was the case with gender, table 32 also indicates that this variable is not correlated with the preference values that were expressed by the sample population.

177 176 Methodological approaches for the protection of cultural heritage in the digital age VIRTUALITY_SCALE INTERACTIVITY_SCALE PRECISION_SCALE VISUAL_CONSISTENCY_ SCALE PRECISION_VISUAL_AG GREGATE_SCALE AUTOMATISM_SCALE Sum of Squares df Mean Square F Sig. Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Table 32: ANOVA table analyzing the influence of the age variable on the actor preferences Education The third independent variable that was retained is the one representing the level of education that characterizes the people that answered our questionnaire. Table 33 shows that for most of the considered dependent variables, the differences observed in the preferences that were expressed by our test subjects are unlikely to be caused by their education level. However, we can notice that education seems correlated to the variation of the preferences regarding interactivity and automatism, with a respective eta squared of.290 and.317 (eta squared tables are available in appendix 3 in the measures of

178 Analysis 177 association section). While this result could be caused by a skew introduced by the limited sample that was used for our tests, it could also reveal a link between their level of education and their preferences towards those particular areas. VIRTUALITY_SCALE INTERACTIVITY_SCALE PRECISION_SCALE VISUAL_CONSISTENCY_SC ALE PRECISION_VISUAL_AGGRE GATE_SCALE AUTOMATISM_SCALE Sum of Squares df Mean Square F Sig. Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Table 33: ANOVA table analyzing the influence of the education variable on the actor preferences In order to assess this issue, further tests accompanied by the formulation of specific hypotheses regarding this phenomenon should be required in order to establish the type of correlation and the pertinence of such relation. However, since our main concern at this stage is testing our variables to find one that could simultaneously explain the variation of all the dependant ones, and since this issue does not directly sit inside the boundaries of our investigation, we

179 178 Methodological approaches for the protection of cultural heritage in the digital age prefer leaving the investigation of this particular aspect, despite its interesting nature, for a further study Technological affinity The fourth considered independent variable is the technological affinity that is displayed by the participants of the survey. We previously raised the question of whether or not any variation of the preferences regarding our specific domain of study, due to is technological nature, might exclusively be explained by the degree of technological affinity featured by our test subjects: being familiar with new technologies and modern 3D applications might, in fact, automatically bring someone to prefer certain approaches independently from their occupation and field of professional activity. VIRTUALITY_SCALE INTERACTIVITY_SCALE PRECISION_SCALE VISUAL_CONSISTENCY_SC ALE PRECISION_VISUAL_AGGRE GATE_SCALE AUTOMATISM_SCALE Sum of Squares df Mean Square F Sig. Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Table 34: ANOVA table analyzing the influence of technological affinity on the actor preferences

180 Analysis 179 While table 34 shows a correlation of this variable with the virtuality scale and the interactivity scale, which is to be expected given our initial concerns, the overall results indicate that our hypothesis about technological affinity does not find an empirical validation, thus suggesting that other factors are involved in the definition of an actor preference profile Domain of activity VIRTUALITY_SCALE INTERACTIVITY_SCALE PRECISION_SCALE VISUAL_CONSISTENCY_ SCALE PRECISION_VISUAL_AG GREGATE_SCALE AUTOMATISM_SCALE Sum Squares of df Mean Square F Sig. Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Table 35: ANOVA table analyzing the influence of the domain of activity variable, or actor class, on the actor preferences

181 180 Methodological approaches for the protection of cultural heritage in the digital age The last independent variable that was tested is the one representing each actor domain of activity, which we also often referred to as its actor class. This variable is the main explicative factor that was hypothesized to guide and inform the expression of each actor preference towards a visual restitution of a heritage item. While this relation could be established intuitively, table 35 suggests that the actor class is indeed connected with each of the dependant variables that define our taxonomical space as introduced in chapter 3. Although it might be argued that for the visual consistency scale this relation is not as strongly established as it is for the other variables, its aggregate could instead be retained for the next step of our study. The presence of this link gives us the chance plot the actor preferences by means of cross tabulations that are infused with meaning while using as axes our explicative and our explained variables. These variables respectively are the actor domain of activity and the five identified continua that were employed for our taxonomy: namely the continuum of virtuality, interactivity, precision, visual consistency and automatism. 5.5 Preference profiles The data that was gathered during our survey, as well as the analysis that was presented in section 5.4, consolidates our original intuition that a correlation exists between the level of virtuality, interactivity, precision, visual consistency and automatism that each actor feels as the most appropriate for a visualization that targets a heritage item and the main domain of activity to which he belongs. As such, in this section, we highlight the individual preference values that have been plotted for each of the studied actor classes. These values are graphically represented by three histograms that depict the preferred level that each actor class considers ideal in regard to the dimensions that were employed to define our taxonomical model. The first two histograms are built by using the data that was gathered from the questionnaires and the data that was collected during the interviews. The third histogram, which will be used to define the final preference profiles, was created by weighting the data from the self administered survey with the data that was recorded during the interviews. Due to the nature our sample, several weighting approaches have been tested in order to produce usable results that could highlight the observed differences between classes and that could be representative of the trends observed in both data sets.

182 Analysis 181 Virtuality Interactivity Precision Visual Consistency Automatism Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Content Holders Table 36: observed majorities in the preference values expressed by each actor class in the data gathered from the questionnaires The final weighing strategy that was implemented employs a ratio of 2/5 for the data that was extracted from the questionnaires. This data is based on the tables and histograms that present the distribution of the preference by actor class for each of the individual taxonomical dimensions: such tables are available in appendix 3 and are summarized by table 36. The chosen weighing strategy also uses a ratio of 3/5 for the data that was collected during the interviews; the final results are rounded at 0.5. Actors active in cultural heritage protection activities Preferred dimensions characterizing a virtual heritage application Necessary Optional unnecessary Content Holders Precision Visual consistency Field Personnel Interactivity Automatism Cultural Intermediaries Content Consumers Digital Intermediaries Visual consistency Visual consistency Precision Interactivity Automatism Interactivity Precision Interactivity Visual consistency Interactivity Automatism Precision Visual consistency Precision Automatism Automatism Table 37: expected results concerning the dimensions characterizing a virtual heritage application favored by each of the actors targeted by the experimental assessment Lastly, we will employ those results to determine to which extent the empirically inferred preference values either stray from, or converge towards, the profiles we expected, as initially introduced in section 2.4 and as summarized by table 37.

183 182 Methodological approaches for the protection of cultural heritagee in the digital age Content Holders The results that were extracted from the questionnaires regarding the actor class that regroups content holders, presents an mild preference towards precision, visual consistency and virtuality, while the data gathered from the interviews primarily indicates a strong inclination towards highh precision as a main requirement, as depicted by figure 43. Figure 43: preference values of content holders based on the data gathered from the questionnaires (left), and the data gathered during the interviews (right) The weighted results, as illustrated by figure 44, definee a profile that considers as a necessary aspect a degree of virtuality that ranges from average to high as well as the presence of highh precision; as optional aspect the presencee of a high visual consistency, while other factors are deemed unnecessary. These preference values are in line with the expected results. Figure 44: preference values of content holders based on the data gathered from the questionnaires weighted against the data collected during the interviews Field Personnel The data extracted from the self administered survey concerning the group that represents filed personnel indicates a preference towards precision, visual consistency and virtuality. However, the data gathered from the interviews suggests a preference for high precision and interactivity, as depictedd by figure 45.

184 Analysis 183 Figure 45: preference values of field personnel based on the data gathered from the questionnaires (left), and the data gathered during the interviews (right) The weighted data, as introduced by figure 46, portraits a profile that regards as necessary elements, whichh must be included in a visualization, an average to high degree of virtuality, the presence of high interactivity and high precision; visual consistency is considered optional and a high automatism, in this instance, is regarded as unnecessary. Figure 46: preference values of field personnel based on the data gathered from the questionnaires weighted against the data collected during the interviews These preference values, while in line with the expected results concerning the level of interactivity, diverge from our a priori analysis about the theoretical requirements that should have characterized this particular actor group. In fact, the empirical results indicate that field personnel displays a much higher preference regarding their favored level of precision, a slightly more pronounced inclination towards visual consistency, and a slightly lower need of high automatism than we previously surmised. The main divergence is their predilection for visualizations that feature a high degree of precision: we can conclude that the actors who belong to the field personnel class not only need to assess a situation as quickly and as interactively as possible in the frame of their activity, but they equally necessitate

185 184 Methodological approaches for the protection of cultural heritagee in the digital age that the visual data at their disposal is as precise as possible. Thesee elements would, in fact, allow them to perform fast and selectively targeted actions in cases where their intervention is required Cultural Intermediaries The information that was plotted from the study of the answers provided by the test subjects that belong to the cultural intermediaries actor group highlights a preference which is directed towards highh degrees of automatism and virtuality. While the data gathered from the interviews confirms such inclination, it equally suggests a preference for high interactivity. Figure 47: preference values of cultural intermediaries based on the data gathered from the questionnaires (left), and the data gathered during the interviews (right) The weighted results, depicted in figure 48, describe a profile that favors visualizationss that feature high degrees of virtuality and automatism while regarding as an optional but welcomed trait the presence of high interactivity. A high Precision and a high visual consistency are not considered in this instance as necessary requirements. Figure 48: preference values of cultural intermediaries based on the data gathered from the questionnaires weighted against the data collected during the interviews

186 Analysis 185 This profile mostly adheres to the expected results, namely concerning the preferred degrees of interactivity and precision. However, some divergences appear in respect to the favored level of automatism, which is slightly higher than previously theorized, and in respect to the preferred degree of visual consistency, which is much lower than initially predicted. The difference that existss between the empirical preferences that cultural intermediaries express regarding visual consistency and the higher importance we surmised it might have in their eyes can be explained by the composition of our sample population. This actor class theoretically regroups several agents whose mission is in nature divulgative, such as journalists, editors or audio- and visual media providers and all the actors that work for educative structures museums. However, in practice, our sample population presented a prevalence of teachers as members of this specific actor group. In their view, given the generally tight deadlines under which teachers are usually required to operatee in order to prepare and organize new educational content, it is likely that they considered that a visualization which is first and foremost targeted to their classes must primarily require the least amount of time and effort to prepare. Such visualizations, since they would be employed mainly as an optional support for their courses, should thereforee only feature an amount of interactivity that is sufficient to engagee their students. Having to develop on top of that a visualization that equally exhibits a high degree of visual consistency would, in their regard, negatively impact the amount of time that is required to prepare it: as such, it would negate the high degree of automatism they deem as an absolutely necessary Content Consumers The analysiss of the data that was gathered from the questionnaires, as well as the information that was collected during the interviews, both presented in figure 49, suggest that content consumers are mainly nclined to prefer visual consistency, precision and virtuality. Figure 49: preference values of content consumers based on the data gathered from the questionnaires (left), and the data gathered during the interviews (right) The profile highlighted by the weighted results, as illustrated by figure 50, is characterized by a preference towards visualizations that feature high degrees of virtuality while exhibiting, as a main requirement, a high level of visual consistency.

187 186 Methodological approaches for the protection of cultural heritagee in the digital age Figure 50: preference values of content consumers based on the data gathered from the questionnaires weighted against the data collected during the interviews Even thoughh optional, highh precision is considered as a positive quality that is welcomed whenever is present, while high interactivity is not regarded as a critical element. These preference values are consistent with the expected results; however the anticipated level of preferred interactivity is slightly lower that initially surmised Digital Intermediaries The last set of histograms that has been produced depicts the preferences of digital intermediaries. The analysis of the questionnaires reveals that this group is primarily concerned by high degrees of virtuality, precision and visual consistency, while the interviewed subjects, as illustrated by figure 51, put lesser emphasis on the importance of high precision whilst favoring a slightly higher level of interactivity. Figure 51: preference values of digital intermediaries based on the data gathered from the questionnaires (left), and the data gathered during the interviews (right) The weighted data, depicted in figure 52, defines a profile that is positively inclined towards visualizations that exhibit high degrees of virtuality, interactivity and visual consistency. In this particularr instance, a high level of precision is

188 Analysis 187 considered optional and a high automatism is not regarded as a element. necessary Figure 52: preference values of digital intermediaries based on the data gathered from the questionnaires weighted against the data collected during the interviews These preferences mostly converge towards the expected values. However, the empirical results slightly diverge from the predicted ones: more importance is given to visual consistency and interactivity while a somewhat weaker interest than we initially envisaged is shown towards precision Scale standardization and final preference values The preference values thatt have been presented in section 5.5 are constructed on a scale that ranges from 0 to 4: 0 indicates a very low preference, 1 a low preference, 2 an average preference, 3 a high preference and 4 a very high preference. The profiles that have been highlighted by those histograms are summarized by table 38. Virtuality Interactivity Precision Visual consistency Automatism Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Content Holders Table 38: preference values by actor class based on the data gathered from the questionnaires weighted against the dataa collected during the interviews, results are rounded at 0.5

189 188 Methodological approaches for the protection of cultural heritagee in the digital age For simplicity s sake, we decided to convert these values, which express a preference in qualitative terms, into a scale that ranges from 0 to 1 to represent as a ratio the intensity of the preference that each actor class exhibits towards each of the specific dimension we used to construct our taxonomical space. Virtuality Interactivity Precision Visual consistency Automatism Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Content Holders Table 39: final preference values by actor class: the results of the questionnaires have been weighted against the results of the interviews and standardized on a 0 to 1 scale Using the recoded values, as presented by table 39, a histogram that displays side by side the final preference profiles for each of the considered actor classes has been produced, as depicted by figure 53. Thesee results summarize the different views that each of the groups that participated to our survey has of the main qualities that a visualization targeted to a heritage item must feature in order to satisfy the specific requirements they respectively feel indissociable from the activity they perform in the frame of the protection of cultural heritage. Figure 53: graphical representation of the final preference values by actor class

190 Analysis 189 This histogram graphically shows the areas that each actor class deems as the most critically essential from their own perspective while it highlights, as introduced by table 40, the differences and the similarities between their respective preferences. Necessary Optional Unnecessary Content Holders Precision Visual consistency Field Personnel Interactivity Precision Visual consistency Cultural Intermediaries Automatism Interactivity Content Consumers Visual consistency Precision Digital Intermediaries Visual consistency Interactivity Precision Interactivity Automatism Automatism Precision Visual consistency Automatism Interactivity Automatism Table 40: empirical weighted results concerning the dimensions characterizing a virtual heritage application favored by each of the actors targeted by the experimental assessment We can further employ these results by coupling the final empirical preference values that emerged during the analysis of our survey with the model that was previously introduced in chapter 3 and which was employed to classify the different approaches that can be used to visualize a heritage item. As such, we can overlay on top of our taxonomical model the final preference profile that characterizes each of the considered actor classes as a set of areas that graphically represent the expression of their preferences. These areas, by encompassing parts of our taxonomical space, highlight the specific approaches and the content creation strategies that are best suited for the type of visualization that each actor prefers. By employing such strategies and approaches as a guideline, it should be possible to ensure that the final results produced by a visualization project targeted at specific user groups would match their specific needs and requirements and would therefore be considered as a useful tool. The expected results concerning the definition of these preference areas and their inclusion in our taxonomical model, which were introduced in section 4.6.6, are summarized by table 41, while the empirical results are presented in table 42.

191 190 Methodological approaches for the protection of cultural heritage in the digital age Actors active in cultural heritage protection activities Preferred type of visualization strategy Preferred type of content creation approach Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Renderings Simulations Digital catalogs Digital panoramas Semantically supplemented 2D images Real time AR simulations Semantically supplemented 3D models Digitally augmented movies Renderings Videogames Stereoscopic visualizations Real time VR simulations Renderings Spline surfaces 3D scanning Image based modeling Digital photography Polygonal modeling Image based modeling Polygonal modeling Spline surfaces Spline surfaces Polygonal modeling Table 41: expected results concerning the visualization strategies and content creation approaches preferred by each of the groups targeted by the experimental assessment Actors active in cultural heritage protection activities Preferred type of visualization strategy Preferred type of content creation approach Content Holders Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Renderings Augmented Movies Real Time VR Semantically supplemented 3D models Real Time VR Semantically supplemented 3D models Digital Panoramas Semantically supplemented 2D images Renderings Augmented Movies Real Time VR Semantically supplemented 3D models Real Time VR Semantically supplemented 3D models Splines 3D scanning Polygonal Modeling Polygonal Modeling Digital Photography Image Based Modeling 3D scanning Splines 3D scanning Polygonal Modeling Polygonal Modeling Table 42: empirical results concerning the visualization strategies and content creation approaches preferred by each of the groups targeted by the experimental assessment

192 Analysis 191 The visualization strategies and content creation approaches that were identified as a match to the preference profiles expressed by each of the actor classes show certain interesting overlaps: the five actor groups are, in fact, clustered in three main sets that share common traits and preferences. The first set contains content holders and content consumers, the second set combines field personnel and digital intermediaries while the third set only features the presence of cultural intermediaries. These overlaps indicate that all of the participants to our study, independently from their assigned actor class, express a clear interest towards semantically supplemented visualizations, may they be 2D, as preferred by field personnel, or 3D, as favored by the other four actor classes. As a concluding observation, we can note that these converging elements suggest that the utilization of polygonal modeling techniques and, to a lesser degree, of semi automatic image based modeling and 3D scanning strategies, whenever applied to semantically supplemented real time 3D VR simulations, should generally provide most multi-purpose visualization projects that are intended for multiple end-user groups with an acceptable trade-off between visual consistency, precision and automatism, while ensuring an adequate degree of interactivity to suit the respective needs and requirements of all the actor groups that are targeted by such projects. 5.6 Summary In this chapter the different steps that have taken in order to recode and weight all the data that was gathered in the frame of our survey for the purpose of building a suitable collection of variables to be used for our analysis have been presented. Proceeding from such data, the strategies that have been chosen to aggregate different sets of variables into continuous scales have been introduced alongside the tests that have been employed to ensure their reliability. Secondly, the results of the univariate descriptive study that was performed on the dependent and independent variables have been plotted as tables and histograms to offer an insight into the structure of our sample population as well as into the distribution of the retained variables. Subsequently, a bivariate analysis has been performed on those variables to isolate the ones that would prove the most pertinent for the construction of the preference profile for each of the examined actor classes. Having identified an appropriate set of relevant variables that could be employed for such a task, a set of histograms has been produced to graphically depict the preference values for each of the user groups that participated in our survey. Finally, the expected outcome of our analysis, as previously surmised on the basis of the theoretical study that was introduced in chapter 3 and chapter 4, had been discussed alongside the presentation of the empirical results while offering an overview of their main convergences and divergences. Lastly, the final preference profiles that were built during the last step of our analysis have been presented accompanied by some concluding remarks on the

193 192 Methodological approaches for the protection of cultural heritage in the digital age perspectives that such results might offer when coupled with our taxonomical model.

194 Results and conclusion 193 Chapter 6. Results and conclusion In this chapter we present a general overview of the results that were obtained in each of the parts that constitute this manuscript. A summary of the elements that have been addressed during the exploration of the concept of cultural heritage and cultural item, as discussed in chapter 2, will be offered alongside a synopsis of the taxonomy that was developed in chapter 3 and the analysis that was performed in chapter 5 and chapter 6. Furthermore, a review of the specific contributions offered by our study will be addressed while accompanied by further considerations on the limitations and the ameliorations that can be applied to our work as well as on the future perspectives and research directions it might encourage. 6.1 Ameliorations It is of paramount importance to note that since the time this study was originally started and by time it was published, several modification have been implemented to the taxonomical model that was presented in chapter 3. These changes were in part a natural evolution of the base concepts that underlie the taxonomical framework presented as part of this work: in fact, the idea that several bi-dimensional models were necessary to account for all the interactions between the considered dimensions, alongside the fact that the 4 plotted axes were not truly independent, was never really satisfactory. In part, the practical implementation of such changes was prompted by the comments that we received during the reviewing process of a paper submitted, and later successfully for accepted for publication, to the ACM Journal of Computing and Cultural Heritage. Figure 54: unified 4D taxonomical space defined by the continua of visual consistency and precision, interactivity, automatism and virtuality visualized as 3D RGB color cube with colored cages to account for the 4th dimension

195 194 Methodological approaches for the protection of cultural heritage in the digital age The article in question [FONI10] presents a condensed version of this thesis and details the core or the theoretical framework that was discussed in chapter 3 alongside some of the data analysis and the empirical results which were presented in chapter 4 and chapter 5. The main criticism that was addressed to the first draft of the document, which reflected the content discussed in this manuscript, was that while we were able to produce a very clear bi-modal assessment, which meant that the concept of the paper would be useful (i.e. a taxonomy of visualization and modeling strategies) the reviewers felt the implementation to be technically flawed in terms of getting the taxonomy right. Two reviewers, one technical and the other from Cultural Heritage, recommend that the triangular taxonomy to be replaced by a multi-dimensional model, pointing out that the idea that one axis was opposed to another instead of being complementary was itself flawed. Precision Interactivity Automatism Virtuality R G B V 1 Restitution drawings Augmented pictures Scale models Physical reconstructions Interactive scale models Live experiments Renderings Digital catalogs Digital panoramas Real time VR simulations Stereoscopic visualizations Computer games Real time AR simulations Augmented movies Semantically supplemented 2D Semantically supplemented 3D Table 43: overview of the 4D coordinate sets associated with each of the considered approaches Following the constructive comments of the editors, a new unified 4D conceptual space has been defined by integrating all the previously presented models into one in order to offer a clearer view of the taxonomical space as it was originally intended, as it is illustrated by figure 54. The resulting 4D taxonomical model is

196 Results and conclusion 195 based on an homogeneous vector field that uses as axes the four identified dimensions of precision (P), interactivity (I), automatism (A) and virtuality (V). The triangular taxonomy previously introduced in chapter 3 was therefore modified and all the 2D triangular plotting spaces that appear in this manuscript were replaced by 3D cubic representations that represent the 4D homogenous vector field featured by our new taxonomical space. Table 43 summarizes with a set of four coordinates the positions that each of the approaches that were considered in this study occupy in the new 4D taxonomical model. As an added benefit, this new 4D model additionally features the explicit possibility to account for an independent variation of all considered dimensions, thus allowing the visualization of all the possible combinations that could exist. Considering the extent of those changes as well as their implications on the content of this manuscript as it is currently organized, it is more than clear that the delicate update of the main body of this work to seamlessly integrate this new and modified content would be at best impractical, and most likely impossible, without resorting to a total rewrite. For these reasons, some of the content explored in the revised version of the paper that was accepted for publication has been included in this concluding chapter while said article has been annexed as appendix Results During the course of our work, a first initial definition of the concept of cultural heritage property, derived from The Hague s convention of 1954, has been introduced and later expanded, due to its originally limited application range, to encompass a larger spectrum of cultural items to which such definition should be applied. In order to apprehend the extension of such application range, the identification of all the additional pertinent dimensions, which could be relevant for the characterization of any potential cultural property, has been undertaken. The distinctive traits of a heritage item, such as size, inherent nature, tangibility and cultural significance have been employed to constitute a general classification of such properties. Furthermore, through the exploitation of such categorization, the scope and focus of our work, in relation to the type of heritage item we are primarily concerned with, and towards which our attention and interest is principally directed in the frame of this study, has been defined as concerning tangible human cultural heritage. Subsequently, the general concept of cultural heritage protection has been discussed and the different interdependent aspects and intervention axes that participate in its definition and application, and whose collaborative efforts and joined contribution represent an essential factor to ensure the effectiveness of such enterprise, have been determined and clarified. As such, the activities that have been identified as related to the concept of heritage protection, such as documentation, investigation, conservation, defense and dissemination, alongside the digitization processes aimed at providing such tasks with new methodological approaches and tools that assist and reinforce their efforts to preserve our common cultural heritage, have been analyzed and examined along with their specificities.

197 196 Methodological approaches for the protection of cultural heritage in the digital age Moreover, amid the different digital tools linked to the newly available information and communication technologies that that can be applied to the different areas of intervention connected to the protection of cultural heritage, the ones that are issued from the fields of computer based visualization, threedimensional representation and interactive exploration have been highlighted as central focus for our work. All the groups and actors involved or participating, whether actively or passively, in any of the axes that reinforce the general protection of cultural heritage, have been identified and classified alongside the relations they feature with all the other intervening parties, the particular needs they display and the aspects of their tasks that can benefit the most from the introduction of new digital tools specifically aimed at easing certain aspects of their missions. Thus, content holders, field personnel, cultural intermediaries, content consumers and digital intermediaries have been defined as general conceptual containers to categorize and regroup such actors. While addressing the creation of a conceptual framework that is capable to provide a classification for of all the possible design choices that could be implemented to visualize a tangible cultural heritage item, a general classification of the different strategies that can be employed to such end, as well as a detailed analysis of specifically chosen techniques, has been presented and discussed. 1. Restitution Drawings 2. Augmented Pictures 3. Scale Models Interactivity Physical Reconstructions 5. Interactive Scale Models 6. Live Experiments 7. Renderings 8. Digital Catalogs 9. Digital panoramas 10. Real time VR simulations 3 1 1a 2 Virtuality Stereoscopic visualizations 12. Videogames 13. Real time AR simulations 14. Digitally augmented movies 15. Semantically supplemented 2D photographs 15a. Semantically supplemented 3D models Figure 55: general overview of the considered visualization strategies as classified using the continuum of interactivity and the continuum of virtuality as axes to define a two dimensional taxonomical space. The approaches relying on digital technologies are highlighted by the gray box

198 Results and conclusion 197 In order to support the creation of a coherent and sufficiently comprehensive taxonomy that is capable to encompass the diverse approaches that could be pertinently employed for such task, specific conceptual dimensions and axes, namely the continua of virtuality, interactivity, precision, visual consistency and automatism, have been defined and analyzed. Such continuous dimensions have subsequently been employed as a base to propose three complementary taxonomical spaces, now replaced by the unified 4D model introduced in section 6.1, to provide an insight into the respective advantages and disadvantages of the considered visualization strategies, as well as to offer a conceptual tool to highlight the situations in which certain technological solutions might best serve specific end user requirements. The first typological classification model that has been discussed is a two dimensional taxonomical space that is defined by the association of the continuum of virtuality with the continuum of interactivity. Such model, as summarized by figure 55, has been divided into four distinct quadrants which have been employed to classify the visualization approaches that exclusively rely on traditional techniques as well as the ones that employ modern digital technologies. For each considered quadrant, as well as for some extreme and mixed cases, a few exemplifying strategies have been presented to illustrate that it is possible, by analogy, to fit in the proposed model other specific approaches that have not been explicitly considered. The second typological classification model that has been introduced, as illustrated by figure 56, combines in a multi dimensional taxonomical space the continua of visual consistency, precision, interactivity and automatism. Such model has been employed to highlight the specific strengths and weaknesses of each of the discussed approaches in respect to each other, as well as the inherent tradeoff mechanisms that have to be considered while implementing such visualization strategies. Interactivity Visual Consistency and Precision a 1. Restitution Drawings 2. Augmented Pictures 3. Scale Models 4. Physical Reconstructions 5. Interactive Scale Models 6. Live Experiments Automatism Figure 56: general overview of the visualization strategies which rely on traditional techniques as classified using the continuum of interactivity, precision, visual consistency and automatism as axes to define a multi dimensional taxonomical space

199 198 Methodological approaches for the protection of cultural heritage in the digital age Finally, a selection of techniques and modeling strategies that are specifically suited for the creation of virtual representations of heritage items by means modern visualization technologies have been introduced and analyzed. Our multidimensional taxonomical model has subsequently been extended with the inclusion of such elements to allow for a more detailed classification of the visualization approaches that specifically rely on such digital technologies. The extended taxonomical space, depicted in figure 57, has been used to schematize the relations that exist between a specific digital visualization approach and a given modeling strategy, as well as to emphasize the pertinence that is exhibited by a given modeling strategy in respect to its eventual application in the frame of a particular visualization project. A Visual Consistency and Precision E a B A. Polygonal modeling B. 3D Scanning C. Image based modeling D. VRML modeling E. Spline surfaces F. Digital photography Interactivity D 13 C 8 9 F Automatism 7. Renderings 8. Digital Catalogs 9. Digital panoramas 10. Real time VR simulations 11. Stereoscopic visualizations 12. Videogames 13. Real time AR simulations 14. Digitally augmented movies 15. Semantically supplemented 2D photographs 15a. Semantically supplemented 3D models Figure 57: general overview of the visualization strategies which rely on digital visualization technologies, as classified using the previously employed multi dimensional taxonomical space, and their position inside the confidence areas defined by the considered content creation approaches

200 Results and conclusion 199 In the fourth chapter of our work the methods which have been selected to experimentally assess the preferences that characterize all the main actors that support with their activities the protection of cultural heritage properties have been overviewed. As such, the main objectives and the research questions that inspire our study have been introduced alongside the determination of the hypotheses and conceptual boundaries that define our survey. Secondly, the analytical tools that have been selected to collect the necessary experimental data and the protocols concerning the data gathering process, by means of a self administered questionnaire and a series of face to face interviews, have been introduced. The dimensions, sub-dimensions and indicators that have been chosen to serve as a base to constitute such tools have been discussed in conjunction with the definition of the dependant and independent variables that have been used for the analysis of the collected data. Virtuality Interactivity Precision Visual consistency Automatism Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Content Holders Table 44: preference values by actor group: the results of the questionnaires have been weighted against the results of the interviews Finally, by employing the answers provided by the participants of our survey, and by using the face to face interviews as a guide to interpret and weight the results of the questionnaire, chapter 5 introduced the construction of five additive scales and their transposition into ordinal metric variables: one for each of the dimensions that define our taxonomical model. The analysis of these variables and the study of the gathered data allowed for the experimental determination of the final preference profiles for each of the groups that were targeted by our study, as presented in table 44 and figure 58.

201 200 Methodological approaches for the protection of cultural heritage in the digital age VIRTUALITY INTERACTIVITY PRECISION VISUAL CONSISTENCY 0 Figure 58: graphical representation of the preference values by actor group The interviews that were carried out as part of our survey, which targeted two experts for each of the considered groups, were further employed to assist in the determination of the dimensions that each group considers necessary, optional or unnecessary to fulfill their primary objectives, as presented in table 45. Necessary Optional Unnecessary Content Holders Precision Visual consistency Field Personnel Interactivity Precision Visual consistency Cultural Intermediaries Automatism Interactivity Content Consumers Digital Intermediaries Visual consistency Visual consistency Interactivity Precision Precision Interactivity Automatism Automatism Precision Visual consistency Automatism Interactivity Automatism Table 45: empirical weighted results concerning the dimensions characterizing a virtual heritage application favored by each of the actors targeted by the experimental assessment The final preference profiles that were highlighted by our analysis have subsequently been plotted as preference volumes in our unified 4D taxonomical space, as depicted by figure 59.

202 Results and conclusion 201 Figure 59: volume of preference in the taxonomical model by actor group. Top left: filed personnel and digital intermediaries. Top right: content consumers and content holders. Bottom center: cultural intermediaries extended volume of preference Concerning the further exploitation of this 4D taxonomical model, we would like to mention that the techniques most commonly used to create digital content for a computer assisted visual representation can easily be integrated in such space as a new layer; this new layer facilitates the presentation of the distinctive qualities and practical usefulness that a considered content creation strategy can offer for any given visualization approach. To summarize the relations that exist between a chosen visualization approach and a given content creation strategy, which are overviewed in table 46, we can define around the extreme vertices of our 4D classification model a volume of affinity, which is presented as a large colored sphere in figure 60. No volume of affinity is defined around vertex [0;0;0] and vertex [1;1;1], since we assume that none of the currently available content creation strategies has the ability maximize simultaneously all considered dimensions. Such volume encompasses a portion of the taxonomical space which contains the visualization approaches that were classified in chapter 3, and that were detailed in table 43, as well as the content creation strategies that are located on the vertices of the model. As such, we can graphically represent the pertinence exhibited by a specific content creation strategy in respect to its eventual application in the frame of a particular approach that is classified inside the taxonomical space.

203 202 Methodological approaches for the protection of cultural heritage in the digital age Descriptive Languages Splines Digital Photography Polygonal Modeling 3D scanning Image Based Modeling 7) Renderings O O O 8) Digital Catalogs O O O 9) Digital Panoramas O O O 10) Real Time VR O 11) Stereo O Visualizations 12) Computer O Games 13) Real Time AR O O O 14) Augmented Movies O O O 15) Semantically suppl. 2D O O 16) Semantically O suppl. 3D Table 46: overview of the pertinence of each content creation strategy for each considered visualization approach The localization of each content creation strategy in the conceptual space as a volume of affinity accounts for its peculiar ability to provide and facilitate, in the frame of a specific visualization approach, the expression of a certain degree of visual consistency, precision, automatism or interactivity. Figure 60: Content creation strategies and their volumes of affinity

204 Results and conclusion 203 Actors active in cultural heritage protection activities Preferred type of visualization strategy Preferred type of content creation approach Content Holders Renderings Augmented Movies Real Time VR Semantically supplemented 3D models Splines 3D scanning Polygonal Modeling Field Personnel Real Time VR Semantically supplemented 3D models Polygonal Modeling Cultural Intermediaries Digital Panoramas Semantically supplemented 2D images Digital Photography Image Based Modeling Content Consumers Renderings Augmented Movies Real Time VR Semantically supplemented 3D models Splines 3D scanning Polygonal Modeling Digital Intermediaries Real Time VR Semantically supplemented 3D models Polygonal Modeling Table 47: empirical results concerning the visualization strategies and content creation approaches preferred by each of the groups targeted by the experimental assessment As a concrete example, we can use real time virtual simulations, which are located near the vertex that is associated with polygonal modeling and which is shared by the affinity volumes that pertain to interactivity and precision. For the creation of the three dimensional models that are to be included in such a simulation, polygonal modeling techniques would be the preferred content creation strategy, while semi automatic image based modeling techniques and VRML could eventually be employed as possible alternatives or complementary approaches. The implementation of these techniques would preserve the high degree of interactivity that characterizes real time simulations while the adoption of a content creation strategy that belongs to another affinity volume, such as 3D scanning techniques for instance, would negate such benefit. 6.3 Applications During the writing of this work and all along its development, since its inception as a general idea, while working on the definition of its conceptual framework as well as in parallel with the finalization of the data gathering and analysis

205 204 Methodological approaches for the protection of cultural heritage in the digital age processes, several occasions have arisen to test and put into practice the general principles that have been outlined in this manuscript. As introduced in section 4.8, the ERATO (ICA3-CT ) and CAHRISMA (ICA3-CT ) projects have been the ground upon which the first experiments on the possible study of inferred actor preferences, which were based on an embryonic version of our taxonomical model, have been attempted. Later, the LIFPLUS (IST ) and EPOCH (IST ) projects, among others, provided further opportunities to implement some of the early results of this study during the preparation of a few selected deliverables. In the first two cases the target audience was composed mainly by cultural intermediaries and content consumers: the main issue at hand was which techniques to use in order to produce the 3D content for the restitutions that were aimed at those groups and which technologies to implement in the final visualization without resorting to the creation of two separate applications. For both these cases we settled for real time VR simulations using polygonal modeling tools to meet the expectations of those parties and not to stray from their main preferences concerning the final use of the application. Additionally, in the second case, the use of pre-computed radiance transfer was included to strengthen the visual consistency which the CAHRISMA deliverables were lacking. The third case featured a similar approach in an augmented reality environment that was mainly targeted at content consumers. The last case, which was aimed at both content holders and content consumers, prompted us to choose a rendered movie as support to maximize the impact of the final visualization. In all those instances valuable information was gathered to improve our conceptual models and our theoretical framework by means of the feedback that our simulations received from the groups they were destined for. 6.4 Contributions The main contributions that emerge from the elements that have been analyzed in the course of our study are fourfold. First, the scope and extension of the application range of the concept of cultural heritage property and of the concept of cultural heritage protection have been determined, defined and discussed in depth. Furthermore, the interdependent nature of the aspects and intervention axes that participate in their application have been clarified and analyzed. Moreover, the contribution that digital tools, as well as modern communication and visualization technologies can offer to the activities that are linked to the protection of cultural heritage have been explored thoroughly. Secondly, all actors involved, either directly or indirectly, in the protection of human tangible cultural heritage have been identified and classified depending on the degree of vicinity that they exhibit with the cultural items that are targeted by their specific activities. Moreover, the interdependent relations that exist between such groups, as well as the stream of information which is related to the cultural items that their actions affect, have been modeled, studied and

206 Results and conclusion 205 presented. Furthermore, an analysis of the specific needs and expectations of such groups, as well as an overview of the tasks that could benefit from the introduction of new digital tools, have been offered. Thirdly, a conceptual framework which offers the capacity to classify all the possible design choices that can be applied to the visualization of a human tangible cultural heritage item, and which provides the ability serve as a general classification tool to characterize and analyze the different strategies that could be employed to achieve such objective, has been proposed. In order to support the creation of a coherent and sufficiently comprehensive conceptual framework that is capable of providing a classification for the possible design choices that target the visualization of cultural heritage items and that encompasses all the diverse approaches that could pertinently be employed for such task, specific conceptual dimensions and axes have also been defined. Such continuous dimensions have been employed as a base to propose a taxonomical space that is aimed at providing an insight into the respective advantages and disadvantages that are offered by such visualization strategies as well as to serve as a conceptual tool to highlight the situations in which certain technological solutions might best serve specific user requirements. Such model has subsequently been employed to assess the strength and weaknesses of each of the discussed approaches in respect to each other, as well as to express the inherent tradeoff mechanisms that exist while considering the implementation of a given visualization strategy in the frame of a particular project. Finally, the analysis of the data that was gathered in the frame of our survey by using questionnaires and face to face interviews, as detailed in chapters 4 and 5, has allowed the construction of a set of profiles that define the preferences that each of the considered actor classes expresses regarding the different approaches that could be used to visualize a heritage item. The study of these profiles has offered the possibility to identify to which extent a convergence exists between such preferences, as it has allowed the identification of some common tendencies that could be prioritized whenever a multipurpose visualization project is simultaneously targeted at several different user groups: the purpose of this set of converging elements is to maximize the presence of the factors that each group considers critical while discarding the elements that are considered incompatible with their needs. To conclude, we believe that the exploitation of the proposed conceptual framework, of the identified preference profiles, as well as of the highlighted common tendencies, as supported by the empirical data, could serve as practical guidelines for the creation of virtual heritage applications that could fulfill the specific needs and the key requirement that are exhibited by each of the main actors connected to the field of heritage protection. By employing such guidelines it should be possible to eliminate the necessity to prepare, through a trial and error process, multiple versions of the same virtual visualization in order to meet the different expectations that characterize each party that is involved, or targeted by, a specific virtual heritage application. Furthermore, we also think that such framework could equally be employed by these user groups as a tool to analyze and to assess their needs in respect to a

207 206 Methodological approaches for the protection of cultural heritage in the digital age visualization project they might consider to develop. Its exploitation would allow the deployment of the most suited approaches and the most appropriate technologies to assist the users that are willing to integrate virtual heritage applications into their working pipelines, as it would maximize the effectiveness of such applications while providing them with the capacity to ensure that their particular needs and specific requirement are satisfied. 6.5 Limitations Despite our aspiration to comprehensively embrace and analyze the different aspects and interrelations that inherently exist whenever a visualization project is applied to the protection of cultural heritage, our work is nonetheless characterized by inevitable limitations in both scope and applicability. First, although other aspects of cultural properties have been introduced and briefly explored, the main limitation of our work is constituted by the fact that our focus has been centered on human tangible cultural heritage. Therefore, even though it would present an undeniable interest, the contribution that modern digital technologies could provide for the protection, for instance, of intangible or natural cultural heritage has not been analyzed in depth. Consequently, this study clearly lacks the ability to be applied as is to such aspects of heritage protection. Similarly, the analysis of the different user groups that has been proposed is confined to the ones that are exclusively active in the specific fields that are related to the protection of human tangible cultural heritage. Secondly, for similar reasons, the fact that our attention has been focused on visualization related techniques, approaches and technologies, has further restricted the scope and application range of our contributions. Our framework has been developed and specifically tailored to assist the analysis of such type of project, thus disallowing its capacity to be employed as is to study the impact and usefulness of other type of approaches, such as, for instance, the ones that are based on audio recordings, acoustic stimulations or tactile feedback. Thirdly, since our primary concern has been to assess and study the impact and the contribution that digital visualization technologies could give in order to support the protection of cultural heritage in the digital age, our taxonomical analysis and the proposed taxonomical spaces, although capable to encompass traditional visualization techniques, were primarily developed to specifically study virtual heritage applications. For such a reason, the applicability and the usefulness of the proposed conceptual models is limited the study of such strategies, precluding the possibility to apply them as is to other pipelines, whose objective is not the visualization of a cultural heritage property. Finally, due to the limited size of the sample population that was employed in the frame of our data analysis, although this factor has been mitigated by the set of interviews that were used to weight our experimental results, it is nevertheless conceivable that certain results might have been different if the same tests would have been run on a larger data pool.

208 Results and conclusion Future Work Given that most of the results of our work are purely theoretical constructs that can be used as conceptual tools to analyze the role of visualization technologies in the frame of heritage related applications, it would be interesting and meaningful as a future research direction to corroborate and test such theoretical tools in a practical working environment, since our experimental validation, though encouraging despite the limited available data, might require a larger sample population to produce more robust results. To achieve such goal, it could be foreseeable to supply a specific interest group with a digitally created visual representation of a heritage item they need to work with, and which was developed by employing the techniques and approaches that are deemed optimal by our theoretical models: subsequently their reactions regarding the suitability of such application to their specific subjectively perceived needs and requirement could be collected. Such experiment would allow an analysis that is based on the presence or the absence of an eventual deviation from the results predicted by our models and would provide significant information regarding the adequacy of our framework to a real world situation. Alternatively, the results of such exercise could offer an indication concerning an eventual inadequacy of the user perceptions regarding their own real needs, due, for instance, to an overestimation of their expectations or an underestimation of the real possibilities that are offered by the currently available digital visualization techniques. A further direction of study which could prove interesting, as previously introduced in section 6.5, would be to directly transpose the elements that were identified in our conceptual framework into a set of more robust indicators to be for the design of a more refined questionnaire that would be distributed to a larger sample population. It would then be possible to analyze more deeply, either quantitatively or qualitatively, the subjective preferences, needs and requirements that the different actors that were identified by our work exhibit towards the visual representation of a heritage item. Such an analysis could provide a more solid experimental ground to identify with a higher degree of certainty the elements that minimize the negative response of each user group while maximizing their positive response to a proposed visualization approach. This would allow a stronger empirical validation of the proposed theoretical guidelines and would permit a better identification of the main elements that the different user groups consider primarily responsible for a successful representation of a cultural heritage property.

209 208 Methodological approaches for the protection of cultural heritage in the digital age

210 List of publications 209 List of publications "A Taxonomy of Visualization Strategies for Cultural Heritage Applications", ACM Journal on Computing and Cultural Heritage, Vol. 3, No. 1, Article 1, June Authors: A. Foni, G. Papagiannakis, N. Magnenat-Thalmann. Animation and Illumination in Ancient Roman Sites, invited talk at the SCCH 2007: Scientific Computing and the Cultural Heritage IWR Workshop, Heidelberg, November Authors: A. Foni, N. Magnenat-Thalmann. Time-Dependant Illumination and Animation of Virtual Hagia-Sophia, The International Journal of Architectural Computing (IJAC), 2007, issue 2, volume 5, pp Authors: A. Foni, G. Papagiannakis, N. Cadi, N. Magnenat-Thalmann. Real Time Animation and Illumination in Ancient Roman Sites, The International Journal of Virtual Reality (IJVR), 2007, 6(1), pp Authors: N. Magnenat-Thalmann, A. Foni, G. Papagiannakis, N. Cadi. Real Time Animation of Ancient Roman Sites, invited paper at GRAPHITE 2006, Kuala Lumpur, December Authors: N. Magnenat-Thalmann, A. Foni, N.Cadi-Yazli. Real-time realistic rendering and subtle body movements for cultural heritage virtual actors, Cultural Convergence & Digital Technology 2006, FHW, Athens, November Authors: G. Papagiannakis, A. Egges, A. Foni, N. Cadi, N. Magnenat-Thalmann. Contextualized Cultural Heritage Simulations in Time-Dependant Virtual Spaces, invited presentation at Computer Space 2006, Sofia, November Authors: A. Foni, N. Magnenat-Thalmann, G. Papagiannakis. Interactive visualization of the Aspendos Historical Site sites using VR technologies, ERATO Symposium, Istanbul, February Authors: A. Foni, G. Papagiannakis, N.Cadi-Yazli, N. Magnenat-Thalmann.

211 210 Methodological approaches for the protection of cultural heritage in the digital age Practical Precomputed Radiance Transfer for Mixed Reality, Virtual Systems and Multimedia 2005, VSMM05, Ghent, Belgium. August Authors: G. Papagiannakis, A. Foni, N. Magnenat-Thalmann. Simulation of the Aspendos theater in Turkey, Virtual Rome Exposition, Rome, October Authors: A. Foni, G. Papagiannakis, N. Magnenat-Thalmann. "Simulating life in ancient sites using mixed reality technology", invited paper at CEIG04, Seville, May Authors: N. Magnenat-Thalmann, G. Papagiannakis, A. Foni, M. Arevalo, N.Cadi-Yazli. Virtual reality brings past to life, BBC news, technology highlights, February (Articolo sul Progetto CAHRISMA). The CAHRISMA Project: 3D Real-Time Virtual Restitution of Sacred Heritage sites in Istanbul, Presentation selected for the CERN booth at the SIS Forum, Geneva, December Authors: A. Foni, G. Papagiannakis, N.Cadi-Yazli, N. Magnenat-Thalmann. "A Virtual Heritage Case Study: A Modern Approach to the Revival of Ancient Historical or Archeological Sites Through Application of 3D Real-Time Computer Graphics", AVIR'03, pp.24-26, September, Authors: A. Foni, G. Papagiannakis, N. Magnenat-Thalmann. "Real-Time recreated ceremonies in VR restituted cultural heritage sites", CIPA XIXth International Symposium, pp , July, Authors: G. Papagiannakis, A. Foni, N. Magnenat-Thalmann. "Virtual Restitution of Endangered Sacred Heritage Sites", YTU Cultural Heritage Workshop, December, Authors: A. Foni, G. Papagiannakis, N. Magnenat-Thalmann. "Virtual Hagia Sophia: Restitution, Visualization and Virtual Life Simulation", UNESCO World Heritage Congress, available on October Authors: A. Foni, G. Papagiannakis, N. Magnenat-Thalmann.

212 List of publications 211 Real-Time Photo Realistic Simulation of Complex Heritage Edifices", VSMM2001, Conference Proceedings, pp , October, Authors: G. Papagiannakis, G. L'Hoste, A. Foni, N. Magnenat-Thalmann.

213 212 Methodological approaches for the protection of cultural heritage in the digital age

214 Bibliography 213 Bibliography ABAWI04 Abawi, D. F., Los Arcos, J. L., Haller, M., Hartmann, W., Huhtala, K. and Träskbäck, M., 2004, A Mixed Reality Museum Guide: The Challenges and its Realization, in the Proceedings of the 10th International Conference on Virtual Systems and Multimedia 2004, Nov Softopia, Okagi City, Japan. ABERT06 Abert, O., Geimer, M. and Muller, S, 2006, Direct and Fast Ray Tracing of NURBS Surfaces Interactive Ray Tracing 2006, IEEE Symposium on Volume, Issue, Sept. 2006, pp ABMAYR05 Abmayr, T., Härtl, F., Reinköster, M. and Fröhlich, C., 2005, Terrestrial laser scanning applications in cultural heritage conservation and civil engineering, Proceedings of the ISPRS Working Group V/4 Workshop, Mestre-Venice, Italy, International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, VOLUME XXXVI, PART 5/W17. ADDISON00 Addison, A.C., 2000, Emerging Trends in Virtual Heritage, IEEE Multimedia, pp ADEL05 Adel, M., 2005, Towards Interactive 3D Representation of Architectural Heritage, International Seminar on the Management of the Shared Mediterranean Heritage (ISMARMED), fifth conference on the modern heritage, Alexandria. AIMATSHAPE08 Aim@shape, 2008, Aim@shape European project Internet site. ALBUQUERQUE00 Albuquerque, A. L. P., Gomes, J. and Velho L., 2000, An Overview on Virtual Sets, in Lecture Notes in Computer Science, Springer Berlin / Heidelberg, Volume 1834/2000, pp

215 214 Methodological approaches for the protection of cultural heritage in the digital age ANSTEY00 Anstey, D., Pape, J. and Sandin, D., 2000, Building a VR Narrative, Proceedings of SPIE, Stereoscopic Displays and Virtual Reality Systems VII (The Engineering Reality of Virtual Reality 2000), vol. 3957, San Jose. APPLE08 Apple developer documentation, 2008, Introduction to QuickTime VR, Apple Inc., /chapter_1_section_1.html ARCO06 ARCO European Project, 2006, project Internet site. D ASTOUS02 D Astous, F., 2002, Guide to questionnaires and surveys, Réseau francophone de sociologie. ATHENA03 Athena Review, 2003, Sir Arthur Evans and the Excavation of the Palace at Knossos, Athena Review, Vol.3, no.3: Minoan Crete, 2003, p.19. AUTODESK08 Autodesk stitcher panorama creation software, AUTODESK08b Autodesk Imagemodeler image based 3D modeling software, AZUMA97 Azuma, R. T., August 1997,"A survey of augmented reality," Presence: Teleoperators and Virtual Environments, vol. 6, no. 4, pp

216 Bibliography 215 AZUMA01 Azuma, R., Baillot, Y., Behringer, R., Feiner, S., Julier, S. and MacIntyre, B., 2001, Recent Advances in Augmented Reality, IEEE Computer Graphics and pplications 21, 6, Nov/Dec 2001, pp BANNO07 Banno, A. and Ikeuchi, K., 2007, Acquisition and rectification of shape data obtained by a moving range sensor, IPSJ Digital Courier, Vol. 3 (2007) pp BECKER98 Becker, L., 1998, Experimental Archaeology, Minnesota State University Mankato, emuseum. logy.html BERALDIN04 Beraldin, J. A., 2004, Integration of Laser Scanning and Close-Range Photogrammetry - The Last Decade and Beyond, XXth International Society for Photogrammetry and Remote Sensing (ISPRS) Congress, Istanbul, Turkey, Commission VII, pp BERNARDINI02 Bernardini, F. and Rushmeier, H., 2002, The 3D Model Acquisition Pipeline, Computer Graphics Forum, Volume 21, Issue 2, June 2002, pp BERNERSLEE01 Berners-Lee, T., Hendler, J. and Lassila, O., 2001, The Semantic Web, Scientific American, May 17, BERTINO05 Bertino, E., Franzoni, S., Mazzoleni, P. and Valtolina S., 2005, Integration of Virtual Reality and Database System Techniques In : Lecture Notes in Computer Science, Springer Berlin / Heidelberg, Volume 3433/2005, pp BLAIS04 Blais, F., 2004, Review of 20 years of range sensor development, Journal of Electronic Imaging, January 2004, Volume 13, Issue 1, pp

217 216 Methodological approaches for the protection of cultural heritage in the digital age BLINN78 Blinn, J., 1978, "Simulation of Wrinkled Surfaces", Computer Graphics, (Proc. Siggraph), Vol. 12, No. 3, August 1978, pp BOEHLER03 Boehler, W. and Marbs. A., 2003, Investigating Laser Scanner Accuracy, Proceedings of XIX International Symposium of CIPA. Antalya, Turkey. BOURKE00 Bourke, P., 2000, Passive Stereographics. BRINKMANN99 Brinkmann, N., 1999, The Art and Science of Digital Compositing, Morgan Kaufmann; 1 edition (Jun ), ISBN-10: BROWN00 Brown, M. S. and Seales, W. B., 2000, Beyond 2D images: effective 3D imaging for library materials, International Conference on Digital Libraries, Proceedings of the fifth ACM conference on Digital libraries, San Antonio, Texas, United States, pp BUR03 Bur, D., Perrin, J. P., Fuchs, A. and Guermonprez, V., 2003, "Laser scanning as a tool for archeological reconstitution: A gallo-roman temple in naix-aux-forges, france", Proceedings of XIX International Symposium of CIPA. Antalya, Turkey. CALORI04 Calori L., Diamanti T., Felicori M., Guidazzoli A., Liguori M.C., Mauri M.A., Pescarin S. and Valentini L., 2004, Databases and Virtual Environments: a Good Match for Communicating Complex Cultural Sites, Proceedings of ACM- SIGGRAPH CAVAZZA03 Cavazza, M., Martin, O., Charles, F., Mead, S.J. and Marichal, X., 2003, Users Acting in Mixed Reality Interactive Storytelling, 2nd International Conference on Virtual Storytelling, 2003.

218 Bibliography 217 CATCH08 CATCH SVC elearning project, CDS08 La cité des sciences, 2008, Paris, France. CHAMPION04 Champion, E., 2004, Heritage Role Playing-History as an Interactive Digital Game. Y. Pisan (ed.), IE2004 Australian Workshop on Interactive Entertainment, Sydney, February 2004, pp CHO97 Cho, C. and Leckenby, J. D., 1997, "Internet-Related Programming Technology and Advertising," in Proceedings of the 1997 Conference of the American Academy of Advertising, M. Carole Macklin, ed., Cincinnati, OH: American Academy of Advertising, 69. CHRISTIEDIGITAL08 Christie Digital Systems, CLIBURN05 Cliburn, D. C., 2005, Portable 3D Stereographic Displays, Proceedings of the 2005 ASCUE Conference, June 12-16, 2005, Myrtle Beach, South Carolina. COHEN98 Cohen, J., Olano, M., Manocha, D., 1998, Appearance-preserving simplification, Proceedings of the 25th annual conference on Computer graphics and interactive techniques, pp COLES73 Coles, J., 1973, Archaeology by experiment, London.

219 218 Methodological approaches for the protection of cultural heritage in the digital age COSMAS02 Cosmas, J., Itegaki, T., Green, D., Grabczewski, E., Weimer, F. et al., 2002, A Novel Multimedia System for Archaeology, Forum international des musées, Museum National d'histoire Naturelle, Paris, 3rd 4th 5th April COZZA55 Cozza, L. and Staccioli, R. A., 1955, Rom, wie es war und wie es ist: illustrierter führer durch Rom, Vision Verlag, Rome. CREATIVEASSEMBLY08 SEGA- Creative Assebly, DEAGUILERA03 de Aguilera, M. and Mendiz, A., 2003, Video games and education: (education in the face of a "parallel school"), Computers in Entertainment (CIE), Volume 1, Issue 1 (October 2003), SECTION: Theories, pp DEBERNARDI70 De Bernardi, F., 1970, Teatri in Asia Minore III. DEBEVEC04 Debevec, P., 2004, The Parthenon, SIGGRAPH 2004 Electronic Theater. DECK13 05 Daedalic Entertainement Deck 13, DEIGHTON96 Deighton, J., 1996, "The Future of Interactive Marketing," Harvard Business Review, 74 (6), pp DELEON00 De Leon, V., Berry, R., 2000, Bringing VR to the Desktop: Are you Game?, in: IEEE Multimedia, April-June, 2000, pp

220 Bibliography 219 DICKEY05 Dickey, M. D., Engaging by design: How engagement strategies in popular computer and video games can inform instructional design, Journal of Educational Technology Research and Development, Volume 53, Number 2 / June, 2005, Springer Boston, pp DIGICULT02 Salzburg Research Forschungsgesellschaft mbh, for the European Commission, DG Information Society - D2, 2002, The DigiCULT Report - Technological landscapes for tomorrow s cultural economy: Unlocking the value of cultural heritage, European Communities, 2002, Luxembourg: Office for Official Publications of the European Communities, ISBN DOBBERT05 Dobbert, T., 2005, Matchmoving: The Invisible Art of Camera Tracking, Sybex; Pap/Cdr edition (February 11, 2005), ISBN-10: DODGSON99 Dodgson, N. A., Moore, J. R. and Lang, S. R., 1999, Multi-view autostereoscopic 3D display, IBC '99 (International Broadcasting Convention), Sep 1999, Amsterdam, pp DONATH03 Donath, D., Petzold, F., Thurow, T. and Weferling, U., 2003, FREAK Consistent surveying and documentation systems as a planning basis for largescale historic buildings, Proceedings of XIX International Symposium of CIPA. Antalya, Turkey. DOULAVERAKIS05 Doulaverakis, C., Kompatsiaris, Y. and Strintzis, M. G., 2005, Ontology-Based Access to Multimedia Cultural Heritage Collections - The REACH Project, EUROCON 2005 Serbia & Montenegro, Belgrade, November 22-24, 2005, Volume 1, pp DRETTAKIS04 Drettakis, G., Roussou, M., Tsingos, N., Reche, A. and Gallo, E., 2004, Imagebased Techniques for the Creation and Display of Photorealistic Interactive Virtual Environments, Proceedings of the Eurographics Symposium on Virtual Environments, Grenoble, France, ACM.

221 220 Methodological approaches for the protection of cultural heritage in the digital age DUDEK04 Dudek, I., Blaise, J.Y., Durand, A., Bénistant, P., 2004, Online 2D/3D Graphic Interfaces using XML "Repurposable" Heritage Contents, 12th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG 2004), Univ. of West Bohemia, Campus Bory, Czech Republic, In Journal of WSCG, Vol. 12, No. 1-3, ISSN ECS-BC05 Biodiversity and Forest Management in British Columbia, 2005, Ecoregion Classification System, Ecosystem Classification in British Columbia. EDUCATIONARCADE08 The Education Arcade, 2008, Revolution. EEC93 Council Directive 93/7/EEC, 1993, on the return of cultural objects unlawfully removed from the territory of a Member State, 15 March 1993, Official Journal L 74, , pp ELHAKIM03 El-Hakim, S.F., Beraldin, J.-A., Picard, M. and Vettore, A., 2003, Effective 3D modeling of heritage sites, Proceedings of the Fourth International Conference on 3D Digital Imaging and Modeling, 2003, pp ELKHOURY06 El-Khoury, N., De Paoli, G., Dorta T., 2006, Digital Reconstruction as a means of understanding a building s history, Education and research in Computer Aided Architectural Design in Europe conference (ecaade) - session 19: digital design education, Thessaly, Volos, Greece. EPOCH08 EPOCH, 2008, EPOCH European network of excellence Internet site. EVANS21 Evans, A.J., , The Palace of Minos - Volumes 1-4.

222 Bibliography 221 EVL08 Electronic Visualization Laboratory, FALQUET01 Falquet, G., Guyot, J., Nerima, L. and Park, S., 2001, Design and Analysis of Virtual Museums, In Proceedings of MW2001 (Museums and the Web), Seattle, March FALQUET03 Falquet, G. and Mottaz Jiang, C.L., 2003, A Framework to Specify Hypertext Interfaces for Ontology Engineering, In Proceedings of the Knowledge Management and Organizational Memories workshop of the 18th IJCAI Conf., Acapulco, Mexico, August FARIN99 Farin, G. E., 1999, Nurbs: From Projective Geometry to Practical Use, AK Peters, Ltd.; 2 Sub edition (March 15, 1999), ISBN-10: FARINELLA06 Farinella, G. M. and Mattiolo, G., 2006, Solutions to 3D Building Reconstruction from Photographs, Eurographics Italian Chapter Conference, FISHER70 Fisher, R. A., 1970, Statistical Methods for Research Workers, 14th edition, New York, Hafner. FIRAXIS08 Take-Two Interactive Software Firaxis, FONI02 Foni, A., Papagiannakis, G., Magnenat-Thalmann, N., 2002, Virtual Hagia Sophia: Restitution, Visualization and Virtual Life Simulation, UNESCO World Heritage Congress, available on

223 222 Methodological approaches for the protection of cultural heritage in the digital age FONI07 Foni, A., Papagiannakis, G., Cadi-Yazli, N., Magnenat-Thalmann, N., 2007, Time-Dependant Illumination and Animation of Virtual Hagia-Sophia, The International Journal of Architectural Computing (IJAC), April 200. FONI10 Foni, A., Papagiannakis, G., Magnenat-Thalmann, N., 2010, "A Taxonomy of Visualization Strategies for Cultural Heritage Applications", ACM Journal on Computing and Cultural Heritage, Vol. 3, No. 1, Article 1. FORTE97 Forte, M, 1997, Virtual Archaeology: Great Discoveries Brought to Life Through Virtual Reality, Thames and Hudson, London. FRISHER04 Frischer, B., 2004, The Digital Roman Forum Project of the Cultural Virtual Reality Laboratory: Remediating the Traditions of Roman Topography, 2nd Italy-United States Workshop: The Reconstruction of Archaeological Landscapes through Digital Technologies, Rome, Italy, November 3-6, GAITATSES01 Gaitatzes, A., Christopoulos, D. and Roussou, M., 2002, Reviving the past: Cultural Heritage meets Virtual Reality, Proceedings of the 2001 conference on Virtual reality, archeology, and cultural heritage (VAST01), Glyfada, Greece, SESSION: Immersive VR theatres and rendering for edutainment, pp GAL02 Gal, V., Le Prado, C., Natkin, S. and Vega, L., 2002, Writing for video games, in Proceedings Laval Virtual (IVRC), GEORGOPOULOS03 Georgopoulos, A., Ioannidis, Ch., Makris, G. N., Karkanis, M., Iliopoulou, Ch., 2003, Large scale monument database design, Proceedings of XIX International Symposium of CIPA. Antalya, Turkey, pp GORDON71 Gordon, G. B., 1971, The UNESCO convention on the illicit movement of art treasures, Harvard International Law Journal, vol. 12, p. 543.

224 Bibliography 223 GRUSSENMEYER99 Grussenmeyer, P., Koehl, M. and Nour el Din, M., 1999, 3D geometric and semantic modeling in historic sites, XVII CIPA International Symposium, October 3-6, 1999, Olinda, Brazil, WG2: Cultural Heritage Information Systems. GUARNIERI04 A. Guarnieri, A., A. Vettore, A., El-Hakim, S. and Gonzo, L., 2004, Digital photogrammetry and laser scanning in cultural heritage Survey, International Archives of ISPRS, N. XXXV. Istanbul, GUIDI05 Guidi, G., Micoli, L., Russo, M., Frischer, B., De Simone, M., Spinetti, A. and Carosso, L., 2005, 3D digitization of a large model of imperial Rome, Fifth International Conference on 3D Digital Imaging and Modeling, 2005, 3DIM 2005, pp GUIDI06 Guidi, G., Frischer, B., Russo, M., Spinetti, A., Carosso, L. and Micoli, L., 2006, Three-dimensional acquisition of large and detailed cultural heritage objects, Machine Vision and Applications, Springer Berlin / Heidelberg, Volume 17, Number 6 / December, 2006, pp HAEMIMONTGAMES08 FX Interactive - Haemimont Games, HALL01 Hall, T., Ciolfi, L., Bannon, L., Fraser, M., Benford, S., Bowers, J., Greenhalgh, C., Hellström, S., Izadi, S., Schnädelbach, H. and Flintham, M., 2001, The Visitor as Virtual Archaeologist: Explorations in Mixed Reality Technology to Enhance Educational and Social Interaction in the Museum, Proceedings of the 2001 conference on Virtual reality, archeology, and cultural heritage (VAST01), Glyfada, Greece, SESSION: Augmented reality and mobile systems, pp HALLE97 Halle, M., 1997, Autostereoscopic displays and computer graphics, Computer Graphics, ACM SIGGRAPH, 31(2), May 1997, pp

225 224 Methodological approaches for the protection of cultural heritage in the digital age HANSON01 Hanson, V. D., 2001 L arte occidentale della Guerra: Descrizione di una battaglia nella Grecia classica, Milano, Garzanti, HARTMAN97 Hartman, J., and Wernecke, J., The VRML 2.0 Handbook, Addison-Wesley Professional, ISBN-10: HEALEY92 Healey, F., Heaton, M. and Lobb, S. J., 1992, Excavations of a Mesolithic Site at Thatcham, Berkshire, Proceedings of the Prehistoric Society 58, pp HOFFMANN99 Hoffmann, V., 1999, Die Hagia Sophia: Bilder aus sechs Jahrhunderten und Gaspare Fossatis Restaurierung der Jahre 1847 bis Bern, Peter Lang. HOLTZMAN98 Holtzman, S., 1998, Digital Mosaics, The aesthetics of Cyberspace, Simon & Schuster, NY. IAKK08 Institutt for Arkeologi, Kunsthistorie og Konservering, 2008, Post use-wear analysis replication experiments, University of Oslo IMAGICA08 Imagica Primatte, INGERSOLL77 Ingersoll, D. W., 1977, Experimental archaeology, Yellen, John E., Macdonald, William (editors), (1977), New York. INGLEHART77 Inglehart, R., 1977, The Silent Revolution, Princeton University Press, 1977.

226 Bibliography 225 IPIX08 Minds-Eye-View Inc., 2008, IPIX immersive photography software. IZENOUR92 Izenour, G., 1992, Roofed Theatres in Classical Antiquity, Yale University Press. JONES02 Jones, J.G. and Christal, M., 2002, The future of virtual museums: On-line, immersive, 3D environments, Austin, TX: Created Realities Group. JONES07 Jones, A., McDowall, I., Yamada, H., Bolas, M. and Debevec, P., 2007, Rendering for an Interactive 360º Light Field Display, International Conference on Computer Graphics and Interactive Techniques, ACM SIGGRAPH 2007 papers, SESSION: Light field & high dynamic range imaging, Article No. 40, San Diego, California. KARSLI03 Karsli F., Ayhan E. and Tunc E., 2003, Building 3d Photo-Texture Model Integrated With GIS For Architectural Heritage Conservation, Proceedings of XIX International Symposium of CIPA. Antalya, Turkey. KARSTEN07 Karsten, L., Eisenbeiss, H., Sauerbier, M., Kupferschmidt, D., Gaisecker, T., Sotoodeh, S. and Hanusch, T., 2007, Combining photogrammetry and laser scanning for the recording and modelling of the Late Intermediate Period site of Pinchango Alto, Palpa, Peru Journal of Archaeological Science 34(10), Amsterdam. KATZ55 Katz, E. and Lazarsfeld, P. F., 1955, Personal Influence, New York: The Free Press. KELLY00 Kelly, D., 2000, Digital Compositing in Depth, Coriolis Group,U.S.; Pap/Cdr edition (April 2000), ISBN-10:

227 226 Methodological approaches for the protection of cultural heritage in the digital age KENDERDINE04 Kenderdine, S., 2004, Avatars at the Flying Palace: Stereographic Panoramas of Angkor Cambodia, 8th International Cultural Heritage Informatics Meeting, Session 18, Archives and Museums Informatics, Berlin, Germany. KIM01 Kim, Y., Kesavadas, T., Paley, S. M., Sanders, D. H., 2001, Real-time animation of King Ashur-nasir-pal II ( BC) in the virtual recreated Northwest Palace, Proceedings of the Seventh International Conference on Virtual Systems and Multimedia (VSMM), 2001, pp KLEVJER01 Klevjer, R., 2001, Computer Game Aesthetics and Media Studies, 15th Nordic Conference on Media and Communication Research. Reykjavik, Iceland. KNOPFLE00 Knöpfele, C Virtual Cathedral of Siena. KRENN06 Krenn, B., Sieber, G. and Petschar, H., 2006, Metadata generation for cultural heritage: creative histories the Josefsplatz experience, EVA Conference 2006, "Digital Cultural Heritage - Essential for Tourism", Vienna. df KULTURFILMS00 Kultur Films, 2000, Lost Treasures of the Ancient World: The Seven Wonders Of The Ancient World, Kultur Films. LANCKORONSKI92 Lanckoronski, K., 1892, Städte Pamphyliens Und Pisidiens II. Prag, Wien, Leipzig. LAZARSFELD44 Lazarsfeld, P. F., Berelson, B. and Gaudet, H., 1944, The people s choice: How the voter makes up his mind in a presidential campaign, New York: Columbia University Press.

228 Bibliography 227 LEE80 Lee, D. T. and Schachter, B. J., 1980, "Two algorithms for constructing a Delaunay triangulation", International Journal of Parallel Programming, Springer Netherlands, Volume 9, Number 3 / June, 1980, pp LEPOURAS01 Lepouras, G., Charitos, D., Vassilakis, C, Charissi, A., Halatsi, L., 2001, Building a VR-Museum in a Museum, Third International Virtual Reality Conference, VRIC2001, Laval, France, May 16-20, LEVOY00 Levoy, M., Pulli, K. and al., 2000, The Digital Michelangelo Project: 3D Scanning of Large Statues, Computer Graphics Proceedings, Siggraph 2000,ACM Press / ACM SIGGRAPH, pp LEVY01 Levy, R.M., 2001, Temple site at Phimai: modeling for the scholar and the tourist, Proceedings of the Seventh International Conference on Virtual Systems and Multimedia (VSMM), 2001, pp LIAROKAPIS02 Liarokapis, F., Petridis, P., Lister, P. F. and White, M., 2002, Multimedia Augmented Reality Interface for E-learning (MARIE), World Transactions on Engineering and Technology Education, Vol.1, No.2, LIBERATI03 Liberati, A.M., 2003, La rappresentazione di Roma antica nel plastico di Gismondi del Museo della Civiltà Romana, in Rome An Ville, maquette et modèle virtuel», Cahiers de la Maison de la Recherche en Sciences Humaines, 33, juin 2003, Caen, pp LIFEPLUS07 Lifeplus European project, 2008, project internet site. LIKERT32 Likert, R.,1932, "A Technique for the Measurement of Attitudes", Archives of Psychology 140, pp

229 228 Methodological approaches for the protection of cultural heritage in the digital age LINDT03 Lindt, I., Herbst, I. And Maercker, M., 2003, Interacting within the Mixed Reality Stage, Workshop Proceedings AVIR 03 / Magnenat-Thalmann, Nadia [Eds.], LIU02 Liu, Y. and Shrum, L. J., 2002, "What is Interactivity and is it Always Such a Good Thing? Implications of Definition, Person, and Situation for the Influence of Interactivity on Advertising Effectiveness," Journal of Advertising, 31 (4), pp LIVATINO06 Livatino, S., Cuciti, D. and Wojciechowski, A., 2006, Virtual Reality and Stereo Visualization in Artistic Heritage Reproduction, In proceedings of 7th Symposium of Virtual Reality, Archeology and Cultural Heritage (VAST), Nicosia, Cyprus, October LOCK03 Lock, G., 2003, Using computers in archaeology. Toward virtual pasts, London: Routledge. LOSCOS03 Loscos, C., Widenfeld, H. R., Roussou, M., Meyer, A., Tecchia, F., Drettakis, G., Gallo, E., Martinez, A. R., Tsingos, N., Chrysanthou, Y., Robert, L., Bergamasco, M., Dettori, A. and Soubra, S., 2003, The CREATE project: mixed reality for design, education, and cultural heritage with a constructivist approach, Proceedings of The Second IEEE and ACM International Symposium on Mixed and Augmented Reality, LOURKAKIS03 Lourakis, M., 2003, "Efficient 3D Camera Matchmoving Using Markerless, Segmentation-Free Plane Tracking", Technical Report 324, Inst. of Computer Science, FORTH. LOUVRE08 Museum du Louvre, 2008, Paris, France.

230 Bibliography 229 LTBC01 Messsage , 2001, Message relatif à la Convention de l UNESCO de 1970 et à la loi fédérale sur le transfert international de biens culturels (LTBC). MAZZOLENI03 Mazzoleni, P., Bertino, E., Valtolina, S., Ferrari, E. and Boeri C., 2003, ViRdB: integrating virtual reality and multimedia databases for customized visualization of cultural heritage, International Conference on Computer Graphics and Interactive Techniques, ACM SIGGRAPH 2003 Sketches & Applications, San Diego, California, SESSION: Social networks, pp MCLAUGHLIN00 McLaughlin, M. L., Sukhatme, G., Shahabi, C., Hespanha, J., Ortega, A. and Medioni, G., 2000, The Haptic Museum, Integrated Media Systems Center, University of Southern California, Los Angeles. MCMILLAN02 McMillan, S.J., 2002, Exploring Models of Interactivity from Multiple Research Traditions: Users, Documents, And Systems, In L. Lievrouw and S. Livingston (Eds.), Handbook of New Media, pp , London: Sage. MESHLAB08 Mashlab, 2008, EPOCH NOE Consortium. MILGRAM94 Milgram, P. and Kishino, F., December 1994, A Taxonomy of mixed reality visual displays, IEICE Transactions on Information Systems, Vol E77-D, No.12. MILGRAM99 Milgram, P. and Colquhoun, H., 1999, "A framework for relating head mounted displays to mixed reality displays", Proceedings of the human factors and ergonomics society, 43rd annual meeting, San Jose, pp MILGRAM99b Milgram, P. and Colquhoun, H., 1999, A Taxonomy of Real and Virtual World Display Integration, Mixed Reality - Merging Real and Virtual Worlds. Y. O. a. H. Tamura, Ohmsha(Tokyo)&Springer Verlag (Berlin): pp.1-16.

231 230 Methodological approaches for the protection of cultural heritage in the digital age MILIUS07 Milius, J., MacDonald, W. J. and Heller, B., 2007, Rome, HD Vision Studios, HBO. MILLER94 Miller, P., Richards, J., 1994, The good, the bad, and the downright misleading: archaeological adoption of computer visualisation. In Computer Applications in Archaeology Edited by Jeremy Huggett & Nick Ryan. Oxford: British Archaeological Reports (Int. Series, 600).pp MITCHELL99 Mitchel, W. L., 1999, Moving the museum onto the Internet: The use of virtual environments in education about ancient Egypt, in Virtual Worlds on the Internet, J.A. Vince and R.A. Earnshaw (eds.), IEEE Computer Society Press, pp MONTI03 Monti, C, Fregonese, L. and Achille, C., 2003, Laser scanner application on complex shapes of architecture. Profiles extraction processing and 3D modelling, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XXXIV-5/W10. NANDI00 Nandi, A. and Marichal, X., 2000, Transfiction, Virtual Reality International Conference, Laval May NMST08 National Museum of Science and Technology, 2008, Milano, Italy. NTA02 National Trust of Australia (Victoria), 2002, Classification Report. OFC98 Rapport du groupe de travail, 1998, Transfert international de bien culturels Convention de UNESCO de 1970 et Convention d UNIDROIT de 1995, Office fédéral de la culture, Berne, Suisse.

232 Bibliography 231 OGLEBY01 Ogleby, C, 2001, Virtual World Heritage: More Than Three Dimensional Models, ISPRS. OHTA99 Ohta, Y., 1999, Mixed Reality: Merging Real and Virtual Worlds, edited by Hideyuki Tamura, ISBN: OIUC08 Virtual museum of the Oriental Institute of the University of Chicago, 2008, O KEFFE89 O Keefe, P. and Prott, L., 1989, Law and the Cultural Heritage, vol. 3: Movement, London/Edinburgh, pp OWEN99 Owen, G. S., 1999, Ray Tracing. PAIANO01 R. Paiano, 2001, "Photographic Virtual Reality and Architectural Heritage", Proceedings of ICHIM September Milan - Italy - Vol.2, pp PAPAGIANNAKIS01 Papagiannakis, G., L'Hoste, G., Foni, A., Magnenat-Thalmann, N., 2001, Real- Time Photo Realistic Simulation of Complex Heritage Edifices, Proceedings of Virtual Systems and Multimedia (VSMM 2001), pp PAPAGIANNAKIS05 Papagiannakis, G., Foni, A., Magnenat-Thalmann, N., 2005, Practical Precomputed Radiance Transfer for Mixed Reality, Proceedings of the Virtual Systems and Multimedia Conference (VSMM05), VSMM Society, pp

233 232 Methodological approaches for the protection of cultural heritage in the digital age PAPAGIANNAKIS05b Papagiannakis, G., Schertenleib, S., O'Kennedy, B., Arévalo-Poizat, M., Magnenat-Thalmann, N., Stoddart, A. and Thalmann, D., 2005, Mixing Virtual and Real scenes in the site of ancient Pompeii, Computer Animation and Virtual Worlds Journal, John Wiley and Sons, Vol. 16, No. 1, pp , February PAQUET01 Paquet, E., El-Hakim, S., Beraldin, J.-A., and Peters, S., 2001, The Virtual Museum: virtualisation of real historical environments and artifacts and threedimensional shape-based searching, Proceedings of the International Symposium on Virtual and Augmented Architectures (VAA'01), Dublin, Ireland. June 21-22, pp PARISI94 Parisi, A. and Pesce, M., 1994, Virtual Reality Markup Language (VRML). wired. com/vrml PARK03 Park, C., Kim, I., Ahn, S. C., Ko, H., Kwon, Y. and Kim, H., 2002, The design and implementation of Kyongju VR Theatre, Proceedings of SPIE, Volume 4756, Third International Conference on Virtual Reality and Its Application in Industry, Zhigeng Pan, Jiaoying Shi, Editors, April 2003, pp PATIAS06 Patias, P., 2006, Cultural Heritage Documentation, International Summer School Digital Recording and 3D Modeling, Aghios Nikolaos, Crete, Greece, April PBC04 Ufficio federale della protezione della popolazione, sezione Protezione dei Beni Culturali, 2004, Protezione dei beni culturali in Svizzera Conservare, Tutelare, Rispettare, DDPS, Berna, Svizzera.

234 Bibliography 233 PESCARIN05 Pescarin, S., Forte, M., Calori, L., Camporesi, Guidazzoli, A. and Imboden, S., 2005, Open Heritage: an Open Source approach to 3d real-time and web-based landscape reconstruction, VSMM2005, Proceedings of XI International Conference on Virtual Systems and Multimedia: Virtual Reality at work in the 21st century, Oct Ghent Belgium, ISBN , (Thwaites Ed.) Budapest 2005, pp PHOTOSYNTH07 Microsoft Live Labs Photosynth software, PIEGL97 Piegl, L. A. and Tiller, W., 1997, The NURBS Book, Springer-Verlag, London, UK, ISBN-10: PITZALIS06 Pitzalis, D., Lahanier, C., Pillay, R., Aitken, G., Russell, A., Smith, D. A., Sinclair, P. A. S., Addis, M. J., Lowe, R., Hafeez, S., Lewis, P. H., Martinez, K. and Schraefel, M. C., 2006, Semantically Exposing Existing Knowledge Repositories: a Case Study in Cultural Heritage, In: The first international conference on Semantics And digital Media Technology (SAMT 2006), 6-8 December 2006, Divani Palace Acropolis, Athensm, Greece. RAFAELI88 Rafaeli, S., 1988, Interactivity: From new media to communication, In R. P. Hawkins, J. M. Wiemann, & S. Pingree (Eds.), ''Sage Annual Review of Communication Research: Advancing Communication Science: Merging Mass and Interpersonal Processes, 16'', pp , Beverly Hills: Sage. RAFAELI07 Rafaeli, S. and Ariel, Y., 2007, "Assessing Interactivity in Computer-Mediated Research", in Joinson, A.N., McKenna, K.Y.A., Postmes, T. and Rieps, U.D. (Eds.) The Oxford Handbook of Internet Psychology, Oxford University Press, Chapter 6, pp REILLY90 Reilly, P., 1990, 'Towards a virtual archaeology", Computer Applications in Archaeology 1990, edited by K. Lockyear and S. Rahtz. oxford: British Archaeological reports (Int. Series 565), pp

235 234 Methodological approaches for the protection of cultural heritage in the digital age REILLY92 Reilly, P., 1992, Three-Dimensional modelling and primary archaeological data. In Archaeology and the Information Age. Edited by P. Reilly and S. Rahtz. London: Routledge, pp REMONDINO06 Remondino, F. and El-Hakim, S., 2006, Image-based 3D Modelling: A Review, The Photogrammetric Record, Volume 21, Issue 115, September 2006, pp REYNOLDS99 Reynolds, P. J., 1999, "The Nature of Experiment in Archaeology", chapter in "Experiment & Design in Archaeology", Oxbow Books 1999, edited by A F Harding. RICCI03 Ricci, R., Fantoni R., Ferri de Collibus, M., Fornetti, G. G., Guarneri, M. and Poggi, C., 2003, High-resolution laser radar for 3D imaging in artwork cataloging, reproduction, and restoration, Proceedings of SPIE -- Volume 5146 Optical Metrology for Arts and Multimedia, Renzo Salimbeni, Editor, October 2003, pp RNM00 Rizzoli New Media, 2000, Gli Etruschi, Palazzo Grassi, Venice, Italy. ROUSSOU99 Roussou, M., 1999, Incorporating Immersive Projection-based Virtual Reality in Public Spaces, in Proceedings of the 3rd International Immersive Projection Technology, Stuttgart, pp SANDERS99 Sanders, D. H., 1999, Virtual Worlds for Archaeological Research and Education," in L. Dingwall, S. Exon, V. Gaffney, S. Laflin, and M. van Leusen, Archaeology in the Age of the Internet, Computer Applications and Quantitative Methods in Archaeology 25th Anniversary Conference. University of Birmingham, British Archaeological Reports #S750, Archaeopress: Oxford. SCHMIDT06 Schmidt, J. U., 2006, Anna s Geheimnis, Computer Space 2006, category: offline multimedia.

236 Bibliography 235 SCOTT00 Scott, R., 2000, Gladiator, DreamWorks SKG, Universal Pictures. SCOTT05 Scott, R., 2005, Kingdom of Heaven, Twentieth Century-Fox Film Corporation. SHAPER78 Schaper, E., 1978, "Fiction and the Suspension of Disbelief", British Journal of Aesthetics 18, pp SHENCHANG95 Shenchang, E. C., 1995, QuickTime VR: an image-based approach to virtual environment navigation, Proceedings of the 22nd annual conference on Computer graphics and interactive techniques, pp SHOUKANG98 Shou-Kang, W., Yu-Fei, H,. and Reinhard, K., 1998, Color Anaglyphs for Panorama Visualizations, TR 19, CITR, Auckland University, NZ, Feb SIDERIS02 Sideris, A., Roussou, M., 2002, Making a new world out of an old one: in search of a common language for archaeological immersive VR representation, proceedings of VSMM02, pp SIGGRAPH05 Siggraph 2005 Course Number 38, 2005, Introduction to Real-Time Ray Tracing. SINCLAIR05 Sinclair, P. A. S., Goodall, S., Lewis, P. H., Martinez, K. and Addis, M. J., 2005, Concept browsing for multimedia retrieval in the SCULPTEUR project, In: The 2nd Annual European Semantic Web Conference, 29 May - 1 June 2005, Heraklion, Crete.

237 236 Methodological approaches for the protection of cultural heritage in the digital age SKRYDSTRUP04 Skrydstrup, M., 2004, A Visionary Idea and a Pragmatic Tool: Making a case for a database listing resolutions to cultural property claims, ICME at Museums and Intangible Heritage ICOM 2004 general conference, Seoul, Korea. October 6, SNYDER07 Snyder, Z., 2007, 300, Warner Bros. Pictures. SORENSEN06 Sorensen, T., 1996, Project: Thoroughbred of the Sea: The trial voyage to Dublin Research Plan, The Viking Museum, Roskilde, Denmark. v2006_uk.pdf SORMANN04 Sormann, M., Reitinger, B., Bauer, J., Klaus, A. and Karner, K., 2004, Fast and detailed 3D reconstruction of cultural heritage, In: proceedings of the International Workshop on Vision Techniques applied to the Rehabilitation of City Centres, Lisbon, Portugal, Oktober SQUIRE03 Squire, K. D. and Jenkins, H., 2003, Harnessing the Power of Games in Education, Insight, 3(1), pp SQUIRE03b Squire, K., 2003, Video games in education, International Journal of Intelligent Games & Simulation, 2. SQUIRE04 Squire, K. and Barab, S., 2004, Replaying history: engaging urban underserved students in learning world history through computer simulation games, International Conference on Learning Sciences, Proceedings of the 6th international conference on Learning sciences, Santa Monica, California, pp

238 Bibliography 237 SRINIVASAN05 Srinivasan, R. and Huang, J., 2005, Fluid ontologies for digital museums, International Journal on Digital Libraries, Springer Berlin / Heidelberg, Issue Volume 5, Number 3 / May, 2005, pp SRM-BC96 Regional Protected Areas Team, 1996, A Protected Areas Strategy for British Columbia: The Prince Rupert Region PAS Report. The protected areas component of B.C.'s Land Use Strategy, Government of British Columbia, Ministry of Sustainable Environment, Skeena Region. STEUER92 Steuer, J., 1992, "Defining Virtual Reality: Dimensions Determining Telepresence," Journal of Communication, 42 (4), pp STRICKER01 Stricker, D., Dähne, P., Seibert, F., Christou, I. T., Almeida, L. and Ioannidis, N., 2001, Design and Development Issues for ARCHEOGUIDE: An Augmented Reality based Cultural Heritage On-site Guide, EuroImage ICAV 3D Conference in Augmented Virtual Environments and Three-dimensional Imaging, Mykonos, Greece, 30 May-01 June SVANAES00 Svanaes, D., 2000, Understanding Interactivity: Steps to a Phenomenology of Human-Computer Interaction, NTNU, Trondheim, Norway. PhD. TAYLOR03 Taylor, J., Beraldin, J. A., Godin, G., Cournoyer, L., Baribeau, R., Blais, F., Rioux, M. and Domey, J., 2003, NRC 3D Technology for Museum and Heritage Applications, Journal of Visualization and Computer Animation, Volume 14, (3), 2003, pp THALMANN04 Magnenat-Thalmann, N., Papagiannakis, G., Foni, A., Arevalo-Poizat, M., Cadi- Yazli, N., 2004, Simulating life in ancient sites using mixed reality technology, Spanish computer graphics conference, invited paper, Seville (CEIG04).

239 238 Methodological approaches for the protection of cultural heritage in the digital age THALMANN07 Magnenat-Thalmann, N., Foni, A., Papagiannakis, G., Cadi-Yazli, N., 2007, Real Time Animation and Illumination in Ancient Roman Sites, The International Journal of Virtual Reality, IPI Press, Vol. 6, No. 1, pp , March TSICHRITZIS91 Tsichritzis, D. and Gibbs, S., 1991, Virtual Museums and Virtual Realities, Proceedings of the International Conference on Hypermedia and Interactivity in Museums, Pittsburgh: Archives & Museum Informatics. UBISOFT08 Ubisoft, UNESCO54 UNESCO, 1954, Convention for the Protection of Cultural Property in the Event of Armed Conflict with Regulations for the Execution of the Convention 1954, The Hague, 14 May URL_ID=13637&URL_DO=DO_TOPIC&URL_SECTION=201.html UNESCO54b UNESCO, 1954, Protocol to the Convention for the Protection of Cultural Property in the Event of Armed conflict 1954, The Hague 14 May URL_ID=15391&URL_DO=DO_TOPIC&URL_SECTION=201.html UNESCO64 UNESCO, 1964, Recommendation on the Means of Prohibiting and Preventing the Illicit Export, Import and Transfer of Ownership of Cultural Property, 19 November URL_ID=13083&URL_DO=DO_TOPIC&URL_SECTION=201.html

240 Bibliography 239 UNESCO70 UNESCO, 1970, Convention on the Means of Prohibiting and Preventing the Illicit Import, Export and Transfer of Ownership of Cultural Property 1970, Paris, 14 November URL_ID=13039&URL_DO=DO_TOPIC&URL_SECTION=201.html UNESCO72 UNESCO, 1972, Convention concerning the Protection of the World Cultural and Natural Heritage 1972, Paris, 16 November URL_ID=13055&URL_DO=DO_TOPIC&URL_SECTION=201.html UNESCO99 UNESCO, 1999, Second Protocol to the Hague Convention of 1954 for the Protection of Cultural Property in the Event of Armed Conflict 1999, The Hague, 26 march URL_ID=15207&URL_DO=DO_TOPIC&URL_SECTION=201.html UNESCO03 UNESCO, 2003, Convention for the Safeguarding of the Intangible Cultural Heritage 2003, Paris, 17 October URL_ID=17716&URL_DO=DO_TOPIC&URL_SECTION=201.html UNESCO08 UNESCO, 2008, World Heritage List. UNESCO08b UNESCO, 2008, IDAMS Statistical analysis software manual, chapter 1, section 3. UNIDROIT95 UNIDROIT, 1995, UNIDROIT Convention on Stolen or Illegally Exported Cultural Objects, Rome, 1995,

241 240 Methodological approaches for the protection of cultural heritage in the digital age UNREAL08 Epic Games, Inc., UKNA07 United Kingdom National Archives, earch4 ULTIMATTE08 Ultimatte Corporation, VALTOLINA06 Valtolina, S., Mazzoleni, P., Franzoni, S. and Bertino, E., 2006, A semantic approach to build personalized interfaces in the cultural heritage domain, Proceedings of the working conference on Advanced visual interfaces, Venezia, Italy, SESSION: Interaction for art, cultural heritage and tourism: short papers, pp VANDENHENGEL07 Van den Hengel, A., Dick, A., Thormählen, T., Ward, B., and Torr, P., 2007, Videotrace: Rapid interactive scene modelling from video, ACM SIGGraph VANOSSENBRUGGEN01 Van Ossenbruggen, J., Hardman, L. and Rutledge, L., 2001, Hypermedia and the Semantic web: A research agenda, CWI technical report INS-R0105, May 31, VERGAUWEN04 Vergauwen, M., Pletinckx, D., Willems, G., Verbiest, F., Van Gool, L. and Helsen, T., 2004, As Time Flies By: Mixed Image and Model-Based Rendering of an Historical Landscape from Helicopter Images, The 5th International Symposium on Virtual Reality, Archaeology and Cultural Heritage (VAST 2004). VERGAUWEN05 Vergauwen, M., Willems, G., Verbiest, F., Van Gool, L.J., Pletinckx, D., 2005, Combining image and model based rendering of an archaeological site, SP:IC(20), No. 9-10, October-November 2005, pp

242 Bibliography 241 VERGAUWEN06 Vergauwen, M. and Van Gool, L., 2006, Web-based 3D Reconstruction Service, Journal of Machine Vision and Applications, Springer Berlin / Heidelberg, Volume 17, Number 6 / December, 2006, pp VIRTUHALL08 Martel, J., 2008, La forteresse viking de Trelleborg, illustrations made for the Journal of Medieval History, VISION ROMA95 Vision Roma S.R.L, 1995, Past and present collection, Dian Publications, Athens, Greece. VLAHAKIS03 Vlahakis, V., "Design and application of the LIFEPLUS Augmented Reality System for continuous, context-sensitive guided tours of indoor and outdoor cultural sites and museums", Proceedings of the 2003 conference on Virtual reality, archeology, and cultural heritage (VAST03). VMC08 The Virtual Museum of Canada, VMJA08 The Virtual Museum of Japanese Arts, VROOM08 Swinburne Space Works, VRWAY08 VRWAY Communications, 2008, Panorama of the 15th century church Santa Maria degli Angioli in Lugano, Switzerland and the frescoes by Bernardino Luini. VRMAG,

243 242 Methodological approaches for the protection of cultural heritage in the digital age WARD01 Ward, A., 2001, The Movie "Gladiator" in Historical Perspective, New England Classical Newsletter, William Wyatt, Jr., May WILTON96 Wilton, A. and Bignamini, I., 1996, Grand Tour. The Lure of Italy in the Eighteenth Century, London. WOJCIECHOWSKI04 Wojciechowski, R., Walczak, K., White, M. and Cellary, W., 2004, Building Virtual and Augmented Reality museum exhibitions, Proceedings of the ninth international conference on 3D Web technology, Monterey, California, SESSION: Production, pp WRIGHT99 Wright, M. W., Watson, G., Dunlop, G. and Middleton, R., 1999, Edinburgh: 200 years of heritage through image-based virtual Environments, VSMM99 Conference, Dundee, September WRIGHT06 Wright, S., 2006, Digital Compositing for Film and Video, Second Edition, Focal Press; 2 edition (May 2, 2006), ISBN-10: X. WYELD07 Wyeld, T. G., Carroll, J., Gibbons, C., Ledwich, B., Leavy, B., Hills, J. and Docherty, M., 2007, Doing Cultural Heritage Using the Torque Game Engine: Supporting Indigenous Storytelling in a 3D Virtual Environment, International Journal of Architectural Computing, Volume 5, Number 2, June 2007, pp (18). YILMAZ02 Yılmaz, N., 2002, Working as an illustrator at Kinet Höyük and Bilkent, Blikent University, department of archeology, history and art, Newsletter No , pp

244 Bibliography 243 ZARA04 Zara, J., 2004, Virtual Reality and Cultural Heritage on the Web, Proceedings of the 7th International Conference on Computer Graphics and Artificial Inteligence (3IA 2004) Limoges, France, pp ZELLE04 Zelle, J. M. and Figura, C., 2004, Simple, Low-Cost Stereographics: VR for Everyone, ACM SIGCSE Bulletin, Volume 36, Issue 1 (March 2004), SESSION: Graphics, pp ZHUKOV98 Zhukov, S., Iones, A., Kronin, G., 1998, Using Light Maps to create realistic lighting in real-time applications, in Proc. of WSCG 98 Conference (Plzen, CZ), pp ZORIN00 Zorin, D., Schröder, P., DeRose, T., Kobbelt, L., Levin, A. and Sweldens, W., 2000, Subdivision for modeling and animation, DSIGGRAPH 2000 course notes,

245 244 Methodological approaches for the protection of cultural heritage in the digital age

246 Appendix 1: ACM JOCCH Paper 245 Appendix 1: ACM JOCCH Paper A Taxonomy of Visualization Strategies for Cultural Heritage Applications Alessandro E. Foni MIRALab, University of Geneva and George Papagiannakis AND Nadia Magnenat-Thalmann MIRALab, University of Geneva In this article we present a general classification of the different approaches that might be employed to constitute a visual representation of a cultural heritage item, including the ones featuring the use of traditional tools as the ones exhibiting the inclusion of modern 2D and 3D digital technologies. In order to establish a coherent taxonomy, specific elements characterizing such approaches will be discussed and employed to assist the definition of a general conceptual framework that will enable the classification of the possible design choices according to their specific characteristics. Finally, a selection of modeling and simulation techniques which are specifically related to the creation process underlying the virtual representation of heritage items by means modern visualization technologies will be explored alongside the main areas of preferences linked to the major actors active in the cultural heritage field. Categories and Subject Descriptors: I.3.8 [Computer Graphics]: Applications, J.5 [Arts and Humanities]: Architecture, J.6 [Computer-Aided Engineering]: Computer-aided design (CAD), I.3.7 [Computer Graphics]: Three-Dimensional Graphics and Realism---Virtual Reality, H.5.1 [Information Interfaces and Presentation]: Multimedia Information Systems---Artificial, augmented, and virtual realities General Terms: Theory, Design, Documentation Additional Key Words and Phrases: Cultural Heritage, Taxonomy, Visualization and modeling strategies

247 246 Methodological approaches for the protection of cultural heritage in the digital age INTRODUCTION The main objective of this article is to propose a coherent overview and qualitative comparison of the state-of-the-art methodological approaches that are available to constitute a 2D or 3D visual representation of a cultural heritage artifact or edifice; namely virtual-digital heritage [Arnold et al 2008] or computational humanities [Cruz-Neira 2003]. However, proceeding to a comprehensive list of all the techniques and technologies that have been used in recent years and explored in the frame of several different projects [Arnold and Geser 2008] [Arnold et al 2008] [Havemann 2008], would prove impractical due to the amount of the combinations and diversity in approaches that have been employed. Hence, to overcome such limitation, a general classification and analytical comparison based on the specific inherent characteristics and criteria that can be identified at varying intensity across all such projects is proposed in this work. The creation of a general taxonomical model constitutes an asset to facilitate the identification of the primary elements that each the groups active in the field of cultural heritage protection regards as being the most important, and upon which it bases its decision to consider a proposed visualization as a successful one or not. Thus, by means of this study, our motivation is to seek to determine which approaches and technologies each of those user-groups perceives as being the most appropriate to satisfy its needs and requirements. As a result, our taxonomy will attempt to establish the necessary conceptual bases to allow the identification of clear and practical methodological guidelines that should match the specificities of the considered end user groups. Such guidelines, if implemented, should then permit an improvement of the effectiveness, usefulness, and impact of future projects related to virtual cultural heritage, as they should allow for shorter prototyping cycles, since the amount of work required to produce a virtual representation matching both the expectations and the needs of a given user group will be optimized. To assist the creation of a coherent taxonomy, the four continuous dimensions, or continua [Milgram and Colquhoun 1999a] [Milgram and Colquhoun 1999b] [Azuma 1997] which have been considered in our study will be introduced in section 1 of the article. These continua are employed in section 2 as conceptual axes to define a taxonomical space, which will allow the classification of the selected approaches, thus, assisting in highlighting their relative differences, similarities, limitations and qualities. This model is employed in section 3 as a tool to perform a general categorization of the visualization strategies targeting cultural heritage items. Furthermore, in section 4 and section 5, the same model it is employed as a base upon which two specific layers are superimposed to allow the exploration of a more detailed classification of digitally-based visualization approaches. The first layer introduces the modeling and simulation strategies specifically related to each of the computer aided visualization techniques considered in our study, while the second layer presents the results of a qualitative survey targeting the preferences expressed by each of the actors involved in the field of cultural heritage. Finally, in section 6, we will present some concluding remarks about out study.

248 Appendix 1: ACM JOCCH Paper Axes and dimensions employed in this study Since we aim at classifying the visualization attempts which are targeting heritage items, the definition of a continuum considering the amount of virtuality [Milgram and Kishino 1994] present in a given visualization is a first dimension to be considered. Another dimension that needs to be considered is the one accounting for the varying degree of interactivity exhibited by a given visual representation. Interactivity, as a conceptual construct, has been widely studied [Rafaeli 1988] [Liu and Shrum2002] [McMillan 2002] [Svanaes 2000] and several definitions have been proposed [Deighton 1996] [Steuer 1992] [Rafaeli and Ariel 2007] [Cho and Leckenby 1997]. In the frame of our study, we consider the degree of interactivity featured by a given visualization as its capacity to contextually offer the possibility to subjectively experience an interactive behavior in a synchronous way, thus enabling the user the opportunity to meaningfully contribute to a given experience or to affect in real time the visualized item. A third multifaceted dimension that shall be considered is the continuum of visual consistency and precision. With the term precision, we do not exclusively refer to the amount of geometrical detail and data that can be fit into a specific visualization: although geometrical detail may represent an important component, we however extend such concept to encompass all the elements that can contribute to enhance the reliability, both historical and scientific, of a given visualization attempt. On the other hand, visual consistency represents the capacity for a visualization attempt to exhibit photorealistic qualities, thus facilitating the establishment of a suspension of the viewer s disbelief [Holtzman 1998] [Schaper 1978]. The fourth, and last, considered dimension is the one constituted by the degree of automatism, or span of the development cycle necessary to complete a visualization project, which accounts for the time and effort necessary for the acquisition and processing of the data required to produce a specific visualization attempt. The 4D taxonomical model built by employing those axes is depicted in figure 1. 2 The 4D Taxonomical model of our work Each of the considered approaches is characterized by a varying intensity of each of the previously introduced four dimensions: it is therefore possible to represent them by a set of four coordinates, each varying from 0 to 1. To account for all possible combinations that exist between the considered continua, that are represented by the four coordinate sets, we chose to employ a 3D cube. Such cube has its origin located in [0;0;0] and its maximum, which is the vertex opposed to the origin, located in [1;1;1]. The points inside the cube thus account for all the possible combinations between the continua associated with the 3 axis adjacent to the cube in a 3D space, which are in our case the degrees of precision (y axis), interactivity (z axis) and automatism (x axis). To represent the 4 th dimension, which accounts for the degree of virtuality, the points inside the cube are colored with a varying shade of gray: 0, or black, being completely real and 1, or white, being completely virtual. However, in order to offer a clearer graphical representation of the model on a 2D support, such as a figure, as to allow an easier interpretation of the positioning of the points relative to each other inside the 3D space, we chose to

249 248 Methodological approaches for the protection of cultural heritage in the digital age employ an isometric view of a 3D RGB color cube. In this model R represents the degree of precision, G the degree of interactivity and B the degree of automatism. Each point is therefore associated to an RGB value which corresponds to its specific location inside the taxonomical space. Furthermore, a cage has been added around each of the points to represent their degree of virtuality: black for prevalently real, grey for mixed, white for prevalently virtual. Figure 1: Taxonomical space defined by the continua of visual consistency and precision, interactivity, automatism and virtuality visualized as 3D RGB color cube with colored cages to account for the 4 th dimension. By employing such model, any considered approach can therefore be compared while being classified as a single point located inside the defined conceptual space: the value of its coordinates in the 4D taxonomical space, presented in table 1, thus account for the capacity of the corresponding approach to exhibit a specific strength along each of the dimensions associated with the model.

250 Appendix 1: ACM JOCCH Paper 249 Precision Interactivity Automatism Virtuality R G B V 1 Restitution drawings Augmented pictures Scale models Physical reconstructions Interactive scale models Live experiments Renderings Digital catalogs Digital panoramas Real time VR simulations Stereoscopic visualizations Computer games Real time AR simulations Augmented movies Semantically supplemented 2D Semantically supplemented 3D Table 1: Overview of the 4D coordinate sets associated with each of the considered approaches. 3 Overview of the identified approaches In this section, we present a brief overview of all the approaches aimed at visualizing a heritage item that have been considered in the frame of our taxonomy and which feature a different combination of the main four criteria we have employed as base of our classification model. In order to highlight the specific properties of each identified approach, we will be referring to the initially introduced 4D classification model which describes the relations liking the continua of: precision (P), interactivity (I), automatism (A), virtuality (V) and the inherent trade-off regarding those dimensions exhibited by each visualization strategy. Our approach can thus be summarized in the following 4D homogeneous vector field Virtual Heritage Taxonomy = [P, I, A, V] which is applied in each of the categories in the following sections. 3.1 Restitution drawings: [P, I, A, V] = [0.1, 0, 0.7, 0] While processing an archeological dig, it is generally a standard procedure and an essential step for the excavation team to define accurate drawings and to execute precise illustrations, as well as to constitute a photographic documentation, of the excavated items as part of the recording process

251 250 Methodological approaches for the protection of cultural heritage in the digital age performed on site [Lock 2003]. The content of these drawings may range from the creation of general maps, to the detailing of architectural and sculptural elements [Izenour 1992], to the definition of plans and sections of the considered items [Lanckoronski 1892], [De Bernardi 1970], [Hoffmann 1999]. 3.2 Augmented pictures: [P, I, A, V] = [0.1, 0.1, 0.8, 0] This kind approach allows for the visual integration and the direct confrontation of selected restitution hypothesis with the present reality. As examples pertaining to his category we can cite, for instance, the augmented reconstructions of Pompeii and Ercolanum prepared by Cozza and Staccioli [1955] as depicted by figure 2 and figure 3, or the visualization attempts developed by Vision Roma [1995], which have been included in books as well as have been published as postcards. Figure 2: Coliseum visualization by Cozza and Staccioli [1955]. Figure 3: Coliseum augmented restitution by Cozza and Staccioli [1955]. 3.3 Scale models: [P, I, A, V] = [0.4;0.3;0.6;0] The approaches aimed at the visualization of a heritage item through the physical reproduction of a scale model generally exhibit a relatively mild visual consistency and precision, depending on the amount of detail enabled the by a chosen scale, the closer being to 1:1, the more detail can be included in the model, and by the type of surface material employed. Moreover, while the development cycle for the creation and finalization of small scaled models, or representing small sized artifacts items, may be reasonable, it is indeed current that the creation of larger physical visualizations, as it is illustrated by Italo Gismondi s plaster models of the ancient city of Rome [Liberati 2003], may require a conspicuous amount of time and effort to achieve completion. 3.4 Physical Reconstructions: [P, I, A, V] = [0.8;0.6;0;0] On-site physical reconstructions, or the process of anastylosis, generally offer the visitors of an archeological site the possibility to visually experience a highly consistent representation of the past appearance of the heritage items. However, the historical accuracy, and global precision, of such strategy may sometimes be questionable if not supported by enough evidence [Athena 2003],

252 Appendix 1: ACM JOCCH Paper 251 as it is the case, for instance, for the interventions executed by Sir Arthur Evans [1921] at the archeological site of Knossos. 3.5 Interactive scale models: [P, I, A, V] = [0.4;0.8;0.5;0] The visualization attempts pertaining to this type of strategy share comparable properties with the ones belonging to the approaches employing static scale models. Such models however present specific moving parts that can be manipulated, animated or activated by the user, thus visually illustrating their role, purpose or construction method. An example of such interactive scale models may be represented the interactive reproductions of the inventions made by Leonardo da Vinci that can be observed in the National Museum of Science and Technology [2008], located in the city of Milan. 3.6 Live experiments: [P, I, A, V] = [0.5;0.7;0.4;0] Live experiments constitute a special case of interactive scale model, where the built replicas are always realized at a scale of 1:1. The highly interactive aspect of this approach resides in the fact that live experiments are intended to be directly experienced by their users through all the phases of their realization. This type of visualization strategy, aimed at the replication of ancient artifacts or known past activities, is best suited and mostly used for hypothesis testing in the research field of experimental archeology [Becker 1998], [Reynolds 1999], [Sorensen 2006]. 3.7 Renderings realized with Computer Graphics: [P, I, A, V] = [0.85;0;0.15;1] To achieve a visually consistent and historically precise 3D virtual restitution of a selected heritage item, different aspects are critical [Foni et al. 2002], such as, for instance, the choice of the appropriate modeling techniques, the preparation of the illumination models, the accuracy applied to the texture and material definition processes, and the reliability of the hypotheses underlying the chosen virtual restitution attempt. As a few examples of such approach, we may cite the radiosity processed movies presenting the virtually restored Church of SS. Sergios and Bacchos, located in Istanbul, made in the frame of the CAHRISMA project [Papagiannakis et al. 2001], or Paul Debevec s photorealistic visualization of the Parthenon in Athens generated by means of high dynamic range image based lighting techniques [Debevec 2004]. 3.8 Digital catalogs: [P, I, A, V] = [0.2;0.15;0.7;1] With the term digital catalog, we refer to the visualization attempts, generally undertaken by museums, to publish in electronic format, for public consultation, the information relative to a specific collection of cultural items, while integrating in an organized manner both the graphical illustrations, either photographic, rendered or hand drawn, and the textual documentations related to each presented artifact in order to facilitate and broaden the access of such collections to a wider audience. The visualization of heritage items though the creation of such digital catalogs is characterized the fact that the real objects

253 252 Methodological approaches for the protection of cultural heritage in the digital age themselves are accessible to the users only as digital proxies. As a few cases exemplifying this approach, we may mention, for instance, the digital catalogs available at the Virtual Museum of Japanese Arts [2008], the Virtual Museum of Canada [2008] or the United Kingdom National Archives [2008]. The rapid advances made in computer graphics and the democratization of high speed internet connections has however encouraged and finally allowed, as introduced by [Tsichritzis and Gibbs 1991], the integration of more complex data [Mitchell 1999] [Paquet et al. 2001], exemplified by figure 4, into such digital spaces, hence introducing a shift in their position in our taxonomical model depending on the type of data featured by such new type of catalogs. 3.9 Digital panoramas: [P, I, A, V] = [0.3;0.4;0.9;1] Digital panoramas have been often employed and widely used since their relatively short and inexpensive development cycle offers comparable degrees in both visual consistency and precision as digital catalogs, while allowing at the same time an increased interactivity. This approach was adopted to realize, for instance, the virtual tour of the Cathedral of Milan and the Basilica of St. Peter in Vatican [Paiano 2001], the visualization of the Santa Maria degli Angioli Church located in Lugano [VRWAY 2008], as well as for the constitution of the virtual environment of the Edinburgh s Royal Mile [Wright et al. 1999]. Additionally, as further examples, we may also cite the augmented visualization of the restituted Byblos theatre presented in El-Khoury et al. [2006], the virtual restorations of the Al-Suq al-faransawy building, located in Alexandria, made by Adel [2005], or the augmented virtual reconstructions targeting the landscape of the Ename site [Vergauwen et al. 2005] depicted in figure 5. Figure 4: Digital Catalogue of the Chapel of Scrovegni based on photogrammetry [Paquet et al. 2001]. Figure 5: Augmented image-based reconstruction of Ename [Vergauwen et al. 2005] Real time virtual reality simulations: [P, I, A, V] = [0.8;0.9;0.4;1] The visualization attempts that rely on real time virtual reality simulations share a few characteristics with computer generated renderings. However, while the finalization of a detailed static rendering usually implies at least several minutes of computing for each produced image, real time visualizations are required to generate and display a minimum of twenty five rendered images per second. The level of interactivity that such applications may exhibit can therefore be remarkable, since real time applications provide their users the possibility to

254 Appendix 1: ACM JOCCH Paper 253 freely manipulate the visualized items and directly explore the modeled virtual worlds by allowing the modification at will of the elements featured in the simulation and the generation of real time visual feedbacks for their performed actions and choices. Among the examples that implement such approach we may cite, for instance, the real time virtual reconstruction of Notre Dame in Paris proposed by De Leon and Berry [2000], the immersive virtual tour of the Sienna cathedral developed by Knöpfele [2000], as well as the Real-time Animation of King Ashur-nasir-pal II presented in Kim et al. [2001]. While real time interactive virtual reality simulations might exhibit fairly photorealistic qualities, as demonstrated by the inhabited real time virtual restitution of the Aspendos Theater presented in Magnenat-Thalmann et al. [2007] and depicted by figure 7, their overall geometrical precision and visual consistency is nonetheless generally limited by the necessity to provide a real time feedback to the user. Therefore, real time visualizations normally have to rely on specific solutions, such as bump mapping [Blinn 1978], normal mapping [Cohen et al. 1998] techniques, pre-computed radiance transfer [Papagiannakis et al. 2005a], as illustrated by figure 6, or light map based strategies [Zhukov et al. 1998], to compensate for the impossibility to interactively display complex and very detailed models. Figure 6:Real-time AR visualization using Precomputed Radiance Transfer by Papagiannakis et al. [2005a]. Figure 7: Real-time VR visualization using area lights by Magnenat-Thalmann et al. [2007] Stereoscopic visualizations [0.7;0.7;0.3;1] The level of virtuality exhibited by such visualizations, as well as their overall perceived visual consistency could be considered as slightly superior to plain real time virtual simulations. The capacity to stimulate in the viewer the perception of real depth implies, in fact, both the availability of an increased amount of digital knowledge concerning the displayed items, to allow the system to correctly generate the stereographic pairs, and an enhancement of the photorealistic qualities exhibited by the visualization. Thus, the capability to trigger the suspension of the user s disbelief is equally strengthened. As a few examples of interactive real time applications that implement such specific techniques we may cite, for instance, the virtual reconstitution of the ancient city of Seorabol site in Korea [Park et al. 2003], the artistic heritage reproductions presented in Livatino et al. [2006], the simulations developed for the foundation of Hellenic world introduced in Gaitatzes et al. [2001], as well as the inhabited real time virtual restitution of the Hagia Sophia edifice as presented in Foni et al. [2007].

255 254 Methodological approaches for the protection of cultural heritage in the digital age 3.12 Computer Games: [P, I, A, V] = [0.6;0.75;0.25;1] Since the primary objective of computer games is generally to entertain and engage the player, and not to inform, teach, educate or disseminate information relative to heritage items [Dickey 2005], the cultural value featured by historically contextualized videogames often constitutes a byproduct of their narrative choices and implemented scenario structures. Thus, the cultural heritage virtual models that might be included in a commercial videogame are usually lacking the sufficient degree of scientific solidity and precision to allow such visualizations [Squire and Barab 2004] [Champion 2004], despite their generally good visual consistency and appeal, to be considered a reliable and exploitable mean to accurately represent a considered heritage item. Recently the application of computer games for education and training has been termed serious games and has attracted extensive attention for their applications in cultural heritage [De Leon and Berry 2000], education and training [de Aguilera and Mendiz 2003], The Education Arcade, MIT, 2004], [Freitas 2007], [Hoffman 2009] Real time augmented reality simulations: [P, I, A, V] = [0.5;0.85;0.25;0.5] Augmented reality systems, which are extensively detailed and described in Azuma [1997], Azuma et al. [2001] and Ohta [1999], enhance the real world with computer generated virtual objects that appear to seamlessly coexist in the same space as the real ones. The simulations that employ such techniques generally exhibit simplified geometrical models to achieve real time capabilities, thus scarifying to a certain degree their overall visual consistency and precision. Moreover, though some attempts have been made to approximate in a real time augmented environment complex illumination models that mimic the lighting featured by their reference real scene [Papagiannakis et al. 2005b], such simulations generally implement approximations to the rendering equation for real-time purposes, thus involving a compromised visual consistency for cultural heritage accuracy. Figure 8: AR museum exhibits [Wojciechowski et al ]. Figure 9: AR-site reconstruction of ancient Olympia [Stricker 2001]. Despite such shortcomings, augmented reality simulations and restitutions can however offer interesting possibilities and useful applications, due to their fairly good interactive capabilities and their capacity to visually stimulate the viewer s engagement, as cultural heritage on-site visualization and exploration tools, as edutainment medium [Lindt et al. 2003], or as visual supports to enhance a

256 Appendix 1: ACM JOCCH Paper 255 museum experience [Wojciechowski et al ], as illustrated by figure 8. As a few examples that exemplify the use of such technologies for the visualization of heritage items we can cite, for instance, the on-site augmented and inhabited real time visualization of a Pompeian Thermopolium [Papagiannakis et al. 2005b], presented in figure 11, [Vlahakis 2003], which was accomplished in the frame of the European project LIFPLUS [2008], the museum guide application targeted at the Guggenheim Museum of contemporary art of Bilbao developed by Abawi et al. [2004], depicted by figure 10, as well as the ARCHEOGUIDE project [Stricker et al. 2001], as presented in figure 9, which was aimed at providing a three dimensional augmented interactive on-site restitution of the Ancient site of Olympia. Figure 10: An MR museum guide [Abawi et al. 2004]. Figure 11: AR life simulation in ancient Pompeii [Papagiannakis et al. 2005b] Digitally augmented movies: [P, I, A, V] = [0.9;0;0.1;0.5] The realization of a fully featured digitally augmented movie is generally a process involving several people over long periods of time and requiring a considerable amount of time and effort to reach completion. Moreover, since the final visual results are basically just standard videos exhibiting the inclusion of digitally created visual effects, their degree of interactivity is therefore extremely limited. As a few examples of digitally augmented high production value movies featuring a visually consistent, but rarely scientifically precise, restitution of ancient heritage sites [Ward 2001], we can cite Rome [Milius et al. 2007], Gladiator [Scott 2000], or Kingdom of Heaven [Scott 2005] Semantically supplemented 2D [P, I, A, V] = [0.2;0.5;0.65;0.5] and 3D [P, I, A, V] = [0.8;0.95;0.25;1] representations The process of augmenting a visual representation with a semantic dimension consist in supplementing the graphical data, such as two dimensional photographic images [Georgopoulos et al. 2003], digital panoramas [Valtolina et al. 2006], or three dimensional virtual models [Calori et al. 2004] with a set structured information, in the form of meaningful attributes and concepts, which provide a context and a meaning associated with the visualized content. Semantically contextualized data might successfully support activities ranging, for instance, from the development of heritage information management tools for museum curators [Doulaverakis et al. 2005], to the creation of customizable guided tours for tourists targeting open heritage sites [Krenn et al. 2006] and closed museums [Mazzoleni et al. 2003]. Additionally, the implementation of

257 256 Methodological approaches for the protection of cultural heritage in the digital age semantically augmented representations might equally constitute a valuable asset and a powerful tool for applications targeting, for instance, the analysis of the documentation related to architectural heritage [Dudek et al. 2004], or the classification and structuring of the material and information implemented in virtual museum projects [Falquet et al. 2001] [Srinivasan and Huang 2005], as well as it might also provide a functional and useful modeling [Grussenmeyer et al. 1999], reconstruction and in situ recording aid [Cosmas et al. 2002] for the documentation of archeological excavation sites. As a few examples witnessing the attempt to integrate semantic descriptions into structured graphical virtual representations of cultural heritage items we may mention, for instance, the SCULPTEUR project [Sinclair et al. 2005], the ARCO project [2008], the reconstruction of the temple of Nasium [Bur et al. 03], as well as the photogrammetric documentation of the Opera House in Hannover [Karsli et al.2003]. 4 Validation method: volumes of preference In this section we introduce the validated results obtained by means of a qualitative survey aimed at determining the volumes of preference exhibited by each of the main actors connected to the field of cultural heritage. 4.1 Groups targeted by the qualitative survey To determine the specific preferences of the main actors present in the field of cultural heritage we considered 5 main groups, each consisting of 10 to 20 persons, as target for the qualitative survey we conducted by means of a multiple choice questionnaire. A schematic view of their priorities and their relation with each other is given in figure 12. Digital Content Consumers Conservation + Defense Transmission + Dissemination Field Personnel Cultural Intermediaries Investigation + Expertise Documentation + Access Content Holders Cultural Heritage Item Figure 12: Relation between the 5 main groups active into heritage protection activities.

258 Appendix 1: ACM JOCCH Paper 257 Additionally, to obtain a complementary expert opinion, and thus gain a further insight on their preferences and priorities, 2 to 3 highly informed specialists, for each of the considered groups, have been invited to participate in a structured face-to-face interview. Content holders represent the group that works in closest proximity with the cultural items by having direct access to them. The main responsibilities of such group are the ones involved with the investigative and documenting activities. Proceeding from a direct observation and analysis of the concerned cultural property they aim at increasing the currently existing knowledge of the past, and the elaboration of objective evaluations concerning the cultural significance of the considered items. Moreover, besides offering expertise and documentation, content holders are further responsible to deploy appropriate measures to guarantee the access of the information and data they produce, in order to assist the efforts of the other intervening groups. As main representative actors belonging to such group, we can then cite historians, archeologists and the parties that technically contribute and support either the documenting or the investigating tasks, such as, for instance, scientific personnel, data collecting personnel, and all the actors which are involved in the deployment of measures aimed at establishing a suitable access and distribution of such data at the intention of the other groups. Field personnel regroups all the actors whose actions rely on the data and documentation previously produced by content holders, and whose interventions are physically directed on the cultural heritage items themselves. Their responsibilities range primarily from the application of conservative interventions to the deployment of defensive and sheltering measures, aimed at the preservation and recovery of the structural integrity of such cultural items. Among the professional categories that can be included into this class we may cite artisans working in the field of restoration, personnel assigned to restructuring activities and groups seeking to preserve and defend the cultural properties from the different threatening elements that could affect their integrity, such as the civil protection, the police, or the fire departments. Cultural intermediaries encompasses all the actors whose actions, proceeding from the collected documentation and from the resulting data produced by cultural holders, are directed toward the establishment of a bridge between the cultural items, and their related cultural aspects, and external actors not necessarily pertaining to categories directly involved into heritage protection activities. The responsibilities that cultural intermediaries shall face are primarily related to the dissemination and transmission of the knowledge, documentation and information that is associated with cultural heritage properties. Amid the different professional categories that are related to this class we can mention journalists, editors or audio-visual media providers, the researchers and groups that produce or furnish the content required to accomplish such disseminative activities, teachers and museums. The actors that usually do not assume any main active role in any of the activities associated with the concept of cultural heritage protection, and that constitute the intended target for the disseminative efforts deployed by cultural intermediaries are regrouped in the category of content consumers. Such group

259 258 Methodological approaches for the protection of cultural heritage in the digital age generally does not feature a strong degree of direct proximity and physical interaction with the cultural items, but tends to apprehend the dimensions characterizing their cultural importance through an indirect or guided relation mediated by cultural intermediaries, without resorting to any direct manipulation of the concerned cultural properties. Content consumers therefore represent the intended audience for the transmission, adaptation, and dissemination of the information produced and provided by content holders for the benefit of nonspecialists. The category that includes digital intermediaries regroups all the actors that are involved in any digitization activity related to cultural heritage properties, and whose main task consists in supporting all the other actors connected to the field of cultural heritage with the development of new methodologies and digital applications that might facilitate their tasks. Their responsibilities therefore vary from the identification of new digital based approaches meeting the specific needs of the abovementioned actors, to the implementation of new technologies and tools featuring a suitable degree of adequacy to the practical requirements exhibited by the groups they seek to assist. In such context, digital intermediaries assume a mediating role between the knowledge they posses of the possibilities offered by the new technologies of information, communication, and visualization, and their potential application to the existing fields linked to the area of cultural heritage protection, 4.2 Preferences expressed by the considered groups The questionnaire we issued to the participants of the qualitative survey featured a battery of questions that have been employed to construct additive scales resulting in five ordinal metric variables: one for each of the dimensions featured by our taxonomical model. Virtuality Interactivity Precision Visual consistency Automatism Field Personnel Cultural Intermediaries Content Consumers Digital Intermediaries Content Holders Table 2: Preference values by actor group: the results of the questionnaires have been weighted against the results of the interviews. It has to be noted that precision and visual consistency, although they are two facets of the same dimension have been measured separately to better study

260 Appendix 1: ACM JOCCH Paper 259 which facet has more impact on a given group. The resulting preference values for each of the considered groups are presented in table 2 and figure VIRTUALITY INTERACTIVITY PRECISION VISUAL CONSISTENCY 0 Figure 13: Graphical representation of the preference values by actor group. The interviews conducted in the frame of our qualitative survey, which targeted two experts for each of the considered groups, were further employed to determine the dimensions that each group considers necessary, optional or unnecessary to fulfill their primary objectives as presented in table 3. Necessary Optional Unnecessary Content Holders Precision Visual consistency Field Personnel Interactivity Precision Visual consistency Cultural Intermediaries Automatism Interactivity Content Consumers Digital Intermediaries Visual consistency Visual consistency Interactivity Precision Precision Interactivity Automatism Automatism Precision Visual consistency Automatism Interactivity Automatism Table 3: Priority order of each of the dimensions featured by the taxonomical model by actor group as derived from the conducted interviews. The final results of the survey have subsequently been included as a new layer in the previously presented taxonomical space. The interviews has been used to

261 260 Methodological approaches for the protection of cultural heritage in the digital age determine which dimensions to consider while plotting the volumes of preference exhibited by each group, depicted in figure 14, while the questionnaires weighted results have been employed to determine the boundaries of each volume of preference along the retained axes. Figure 14: Volume of preference in the taxonomical model by actor group. Top left: filed personnel and digital intermediaries. Top right: content consumers and content holders. Bottom center: cultural intermediaries extended volume of preference. It has to be noted that groups exhibiting the same volume of preference have been depicted together in the same diagram, and that in one case, namely for the volume of preference associated with cultural intermediaries, no approach was directly inside the plotted area. In such case we extended the volume along the pertinent axes until some approaches were included in the new extended volume of preference. 5 Results In this section, the role played by the most diffused techniques employed to create the digital content for a computer assisted visual representation is integrated in the previously presented taxonomical model as a new layer. To facilitate the presentation of the distinctive qualities and practical usefulness of the considered content creation strategies a general conceptual categorization, as introduced by figure 15, is proposed. The localization of each content creation strategy in the conceptual space accounts for its peculiar ability to provide and facilitate, in the frame of a specific visualization approach, the expression of a certain degree of visual consistency, precision, automatism or interactivity. Each

262 Appendix 1: ACM JOCCH Paper 261 given combination of dimensions is visually represented as implying constraining and absolute repercussions in respect to the trade-off they imply. Therefore, the presented strategies can support a given visualization project with the exclusive capacity to either enhance the intensity exhibited by a single dimension, or simultaneously assist the full expression of its two adjacent dimensions. For example the use of spline surfaces can only maximize precision and visual consistency, while image based modeling techniques can maximize both interactivity and automatism. Figure 15: Content creation strategies and their volumes of affinity. Such choice allows a simplified graphical localization and classification of each of the considered content creation strategies, which are hence located on the vertices of the conceptual model, with the exclusion of [0;0;0] and [1;1;1], since we assume that none of the currently available content creation strategies has the ability maximize simultaneously all considered dimensions. To summarize the relations existing between a chosen visualization approach and a given content creation strategy, overviewed in table 4, we define around the extreme vertices of the classification model a volume of affinity, which is presented as a large colored sphere in figure 15. Such volume encompasses both a portion of the taxonomical space, which contains the visualization approaches classified in section 2 and detailed in section 3, and the content creation strategies that are located on the vertices of the model. As such, we can graphically represent the pertinence exhibited by a specific content creation strategy in respect to its eventual application to a particular approach that is classified inside the taxonomical space.

263 262 Methodological approaches for the protection of cultural heritage in the digital age Descriptive Languages Splines Digital Photography Polygonal Modeling 3D scanning 7) Renderings O O O 8) Digital Catalogs 9) Digital Panoramas 10) Real Time VR 11) Stereo Visualizations 12) Computer Games 13) Real Time AR 14) Augmented Movies 15) Semantically suppl. 2D 16) Semantically suppl. 3D O O O O O O O O O O O O O O O O O Image Modeling O Based Table 4: Overview of the pertinence of each content creation strategy for each considered visualization approach. As a concrete example, we might consider real time virtual simulations, overviewed in section 3.10, which are located near the vertex associated with polygonal modeling and shared by the affinity volumes pertaining to interactivity and precision. Thus, for the constitution of the three dimensional models to be included in the simulation, polygonal modeling techniques would be the preferred content creation strategy, while image based and VRML modeling techniques could eventually be employed as possible alternatives or complementary approaches. In fact, the implementation of such techniques would indeed preserve the high degree of interactivity characterizing real time simulations, while the adoption of a content creation strategy residing into another affinity volume, such as 3D scanning techniques for instance, would negate such feature. 6 Conclusions In order to support the creation of a coherent and sufficiently comprehensive conceptual framework capable of both providing a classification for the possible design choices targeting the visualization cultural heritage items, and encompassing the diverse approaches that could pertinently be employed for such task, specific conceptual dimensions and axes, have been defined. Such continuous dimensions have been employed as a base to propose a four dimensional taxonomical space, whose aim is to both provide an insight on the respective advantages and disadvantages of the considered visualization strategies, and to offer a conceptual tool to highlight the situations in which certain technological solutions might best serve specific user requirements.

264 Appendix 1: ACM JOCCH Paper 263 Such model has been employed to assess the specific strength and weaknesses of each of the previously discussed approaches in respect to each other, as well as to express the inherent tradeoff mechanisms that have to be pondered while considering the implementation of specific visualization strategies. Furthermore, by employing the results of a qualitative survey, specific volumes of preference have been defined as a layer in the presented conceptual framework. This layer has then been used as a tool to assess the specific needs exhibited by each of the main actors connected to the field of cultural heritage in respect to a visualization project they might consider realizing. Its implementation hence highlights the most suited approaches and the most appropriate technologies to assist the creation of virtual heritage applications which could provide an optimal effectiveness while ensuring the fulfillment of the key requirements expressed by such actors. Finally, a selection of techniques and modeling strategies which are specifically employed for the constitution of virtual representation of heritage items by means modern visualization technologies have been introduced as a second layer in the presented conceptual framework. This second layer has been used to schematize and highlight the relations existing between a specific digital visualization approach and a given modeling strategy, in order to emphasize the pertinence exhibited by a given modeling strategy in respect to its eventual application in the frame of a particular visualization approach. However, it is important to stress that there is no single definitive or ideal solution to simultaneously satisfy all the particular needs exhibited by such actors: ultimately, the optimal choice regarding the strategy to be implemented inherently depends on the resources and the know-how available to them, the specific objectives they want to fulfill, and the type of heritage item being visualized. Nonetheless, by referring to the distribution of the visualization strategies and to the volumes of confidence presented in this article we conclude that the utilization of polygonal modeling techniques and, to a lesser degree, of image based modeling strategies, should generally provide most visualization projects with an acceptable trade-off between visual consistency, precision and automatism, while ensuring an adequate degree of interactivity to suit the requirements of most virtual heritage applications.

265 264 Methodological approaches for the protection of cultural heritage in the digital age REFERENCES ABAWI, D. F., LOS ARCOS, J. L., HALLER, M., HARTMANN, W., HUHTALA, K. AND TRÄSKBÄCK, M A Mixed Reality Museum Guide: The Challenges and its Realization. In Proceedings of the 10th International Conference on Virtual Systems and Multimedia, Okagi City, Japan, November 2004, IOS Press, ADEL, M Towards Interactive 3D Representation of Architectural Heritage. In Proceedings of the International Seminar on the Management of the Shared Mediterranean Heritage, 5th Conference on Modern Heritage, Alexandria, Egypt, March 2005, Ismarmed, electronic version available at: ATHENA Review Sir Arthur Evans and the Excavation of the Palace at Knossos. Athena Review, Vol.3, no.3, 19. ARCO European Project Arco Project, Project Internet Site: ARNOLD D., GESER, G EPOCH Research Agenda for the Applications of ICT to Cultural Heritage. EPOCH (2008) vol. ISBN , pp ARNOLD, D., DAY, A., GLAUERT, J., HAEGLER, S., JENNINGS, V., KEVELHAM, B., LAYCOCK R, MAGNENAT-THALMANN, N., MAIM, J., MAUPU, D., PAPAGIANNAKIS, G., THALMANN, D., YERSIN, B., RODRIGUEZ-ECHAVARRIA, K Tools for Populating Cultural Heritage Environments with Interactive Virtual Humans. EPOCH Conference on Open Digital Cultural Heritage Systems, 1-7. AZUMA, R. T A survey of augmented reality. Presence: Teleoperators and Virtual Environments, vol. 6, no. 4, AZUMA, R., BAILLOT, Y., BEHRINGER, R., FEINER, S., JULIER, S. AND MACINTYRE, B Recent Advances in Augmented Reality. IEEE Computer Graphics and applications 21, 6,

266 Appendix 1: ACM JOCCH Paper 265 BECKER, L Experimental Archaeology. emuseum of the Minnesota State University, Mankato, logy.html BLINN, J Simulation of Wrinkled Surfaces. In Proceedings of the 5th annual conference on Computer graphics and interactive techniques SIGGRAPH 78, August 1978, New York, USA, 1978, ACM Press, BUR, D., PERRIN, J. P., FUCHS, A. AND GUERMONPREZ, V Laser scanning as a tool for archeological reconstitution: A gallo-roman temple in naixaux-forges, France. In Proceedings of the XIXth International Symposium of the International Committee for Architectural Photogrammetry (CIPA), Antalya, Turkey, September CALORI L., DIAMANTI T., FELICORI M., GUIDAZZOLI A., LIGUORI M.C., MAURI M.A., PESCARIN S. AND VALENTINI L DATABASES AND VIRTUAL ENVIRONMENTS: A GOOD MATCH FOR COMMUNICATING COMPLEX CULTURAL SITES. IN ACM SIGGRAPH 2004 EDUCATORS PROGRAM, LOS ANGELES, CALIFORNIA, AUGUST 2004, T. LONGSON, ED. SIGGRAPH '04. ACM PRESS, NEW YORK, NY, 30. CHAMPION, E Heritage Role Playing-History as an Interactive Digital Game. In Proceedings of the Workshop on Interactive Entertainment, Sydney, NSW, Australia, February 2004, Y. PISAN, CHO, C. AND LECKENBY, J. D Internet-Related Programming Technology and Advertising. In Proceedings of the Conference of the American Academy of Advertising, 1997, M. CAROLE MACKLIN, Cincinnati, OH: American Academy of Advertising, 69. COHEN, J., OLANO, M., MANOCHA, D Appearance-preserving simplification. In Proceedings of the 25th Annual Conference on Computer Graphics and interactive Techniques SIGGRAPH '98. ACM Press, New York, NY, COSMAS, J., ITEGAKI, T., GREEN, D., GRABCZEWSKI, E., WEIMER, F. ET AL A Novel Multimedia System for Archaeology. In Proceedings of the Forum International des Musées. Museum National d'histoire Naturelle, Paris, France, April COZZA, L. AND STACCIOLI, R. A Rom, wie es war und wie es ist: illustrierter führer durch Rom. Vision Verlag, Rome, Italy.

267 266 Methodological approaches for the protection of cultural heritage in the digital age CRUZ-NEIRA Computational humanities: the new challenge for VR. IEEE Computer Graphics and Applications (2003) vol. 23 (3), pp DE AGUILERA, M. AND MENDIZ, A Video games and education: education in the face of a "parallel school". In Computers in Entertainment (CIE), Volume 1, Issue 1, October 2003, ACM Press, New York, NY, USA, DE BERNARDI, F Teatri Classici in Asia Minore III, L'Ermadi Bretschneider, ISBN: DEBEVEC, P The Parthenon. SIGGRAPH 2004 Electronic Theater, August DEIGHTON, J The Future of Interactive Marketing. Harvard Business Review, 74 (6), DE LEON, V., BERRY, R Bringing VR to the Desktop: Are you Game?.IEEE Multimedia, April-June, 2000, DICKEY, M. D Engaging by design: How engagement strategies in popular computer and video games can inform instructional design. Journal Educational Technology Research and Development, Volume 53, Number 2 / June, 2005, DOULAVERAKIS, C., KOMPATSIARIS, Y. AND STRINTZIS, M. G Ontology-Based Access to Multimedia Cultural Heritage Collections - The REACH Project. In Proceedings of the International Conference on "Computer as a tool" EUROCON, Volume 1, Serbia & Montenegro, Belgrade, November 2005, DUDEK, I., BLAISE, J.Y., DURAND, A., BÉNISTANT, P Online 2D/3D Graphic Interfaces using XML "Repurposable" Heritage Contents. Journal of the 12th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG 2004), Vol. 12, No. 1-3, EL-KHOURY, N., DE PAOLI, G., DORTA T Digital Reconstruction as a means of understanding a building s history. In Proceedings of the 24th ecaade Conference, Volos, Greece, September 2006, ISBN ,

268 Appendix 1: ACM JOCCH Paper 267 EVANS, A.J The Palace of Minos: a comparative account of the successive stages of the early Cretan civilization as illustrated by the discoveries at Knossos - Volumes 1-4, Macmillan, London, UK. FALQUET, G., GUYOT, J., NERIMA, L. AND PARK, S DESIGN AND ANALYSIS OF VIRTUAL MUSEUMS. IN PROCEEDINGS OF MW2001 (MUSEUMS AND THE WEB), SEATTLE, MARCH 2001, IGI PUBLISHING, HERSHEY, PA, USA, FIRAXIS, Take-Two Interactive Software, 2K Games 2001.Sid Meier's Civilization III, Internet site: FONI, A., PAPAGIANNAKIS, G., MAGNENAT-THALMANN, N Virtual Hagia Sophia: Restitution, Visualization and Virtual Life Simulation. In electronic Proceedings of the UNESCO World Heritage Congress, electronic version available at: FONI, A., PAPAGIANNAKIS, G., CADI-YAZLI, N., MAGNENAT-THALMANN, N Time-Dependant Illumination and Animation of Virtual Hagia-Sophia. The International Journal of Architectural Computing (IJAC), June 2007, FREITAS, S., Learning in immersive worlds. Bristol Joint Information Systems Committee. report_v3.pdf (last accessed December 1, 2009) GAITATZES, A., CHRISTOPOULOS, D. AND ROUSSOU, M Reviving the past: Cultural Heritage meets Virtual Reality. In Proceedings of the 2001 conference on Virtual reality, archeology, and cultural heritage (VAST01), Glyfada, Greece, November 2001, ACM Press, New York, NY, USA, GEORGOPOULOS, A., IOANNIDIS, CH., MAKRIS, G. N., KARKANIS, M., ILIOPOULOU, CH Large scale monument database design. In Proceedings of the XIXth International Symposium of the International Committee for Architectural Photogrammetry (CIPA), Antalya, Turkey, September 2003,

269 268 Methodological approaches for the protection of cultural heritage in the digital age GRUSSENMEYER, P., KOEHL, M. AND NOUR EL DIN, M D geometric and semantic modeling in historic sites. In Proceedings of the XVII International Symposium of the International Committee for Architectural Photogrammetry (CIPA), Olinda, Brazil, October 1999, electronic version available at: HAVEMANN, S., SETTGAST, V., FELLNER, D., WILLEMS, G., GOOL, L., MULLER, G., SCHNEIDER, M., KLEIN, R The Presentation of Cultural Heritage Models in Epoch. EPOCH Conference on Open Digital Cultural Heritage Systems, HELLSTRÖM, S., IZADI, S., SCHNÄDELBACH, H. AND FLINTHAM, M The Visitor as Virtual Archaeologist: Explorations in Mixed Reality Technology to Enhance Educational and Social Interaction in the Museum. In Proceedings of the conference on Virtual reality, archeology, and cultural heritage (VAST), Glyfada, Greece, November 2001, ACM Press, New York, NY, USA, HOFFMANN, V Die Hagia Sophia: Bilder aus sechs Jahrhunderten und Gaspare Fossatis Restaurierung der Jahre 1847 bis 1849, Peter Lang, Bern, Switzerland. HOFFMANN, L Learning through games. Communications of the ACM HOLTZMAN, S Digital Mosaics, The aesthetics of Cyberspace. Simon & Schuster, NY. IZENOUR, G Roofed Theatres in Classical Antiquity. New Haven: Yale University Press. KARSLI F., AYHAN E. AND TUNC E Building 3d Photo-Texture Model Integrated With GIS For Architectural Heritage Conservation. In Proceedings of the XIXth International Symposium of the International Committee for Architectural Photogrammetry (CIPA), Antalya, Turkey, September KIM, Y., KESAVADAS, T., PALEY, S. M., SANDERS, D. H Real-time animation of King Ashur-nasir-pal II ( BC) in the virtual recreated Northwest Palace. In Proceedings of the Seventh International Conference on Virtual Systems and Multimedia (VSMM), Berkley, USA, October 2001, IEEE Computer Society, Washington, DC, USA, KNÖPFELE, C Virtual Cathedral of Siena, Project Internet Site:

270 Appendix 1: ACM JOCCH Paper 269 KRENN, B., SIEBER, G. AND PETSCHAR, H Metadata generation for cultural heritage: creative histories the Josefsplatz experience. In Proceedings of the EVA (Electronic Information, the Visual Arts and Beyond) conference, Vienna, Austria, August 2006, electronic version available at: df LANCKORONSKI, K Städte Pamphyliens Und Pisidiens II. German ed., Vienna. LIBERATI, A.M La rappresentazione di Roma antica nel plastico di Gismondi del Museo della Civiltà Romana. In Rome An Ville, maquette et modèle virtuel, Cahiers de la Maison de la Recherche en Sciences Humaines, juin 2003, Caen, LIFEPLUS European project Lifeplus Project, Project Internet Site: LINDT, I., HERBST, I. AND MAERCKER, M Interacting within the Mixed Reality Stage. In Proceedings of the first research workshop on augmented virtual reality (AVIR), Geneva, Switzerland, September 2003, Eds. Magnenat- Thalmann, Nadia. LIU, Y. AND SHRUM, L. J What is Interactivity and is it Always Such a Good Thing? Implications of Definition, Person, and Situation for the Influence of Interactivity on Advertising Effectiveness. Journal of Advertising, 31 (4), LIVATINO, S., CUCITI, D. AND WOJCIECHOWSKI, A Virtual Reality and Stereo Visualization in Artistic Heritage Reproduction. In Proceedings of 7th Symposium of Virtual Reality, Archeology and Cultural Heritage (VAST), Nicosia, Cyprus, October LOCK, G Using computers in archaeology. Toward virtual pasts. Routledge, London, UK. MAGNENAT-THALMANN, N., FONI, A., PAPAGIANNAKIS, G., CADI-YAZLI, N Real Time Animation and Illumination in Ancient Roman Sites. The International Journal of Virtual Reality, IPI Press, Vol. 6, No. 1, March 2007,

271 270 Methodological approaches for the protection of cultural heritage in the digital age MAZZOLENI, P., BERTINO, E., VALTOLINA, S., FERRARI, E. AND BOERI C ViRdB: integrating virtual reality and multimedia databases for customized visualization of cultural heritage. In Proceedings of the International Conference on Computer Graphics and Interactive Techniques SIGGRAPH 03, San Diego, California, July 2003, ACM Press, New York, NY, USA, 1-1. MCMILLAN, S. J Exploring Models of Interactivity from Multiple Research Traditions: Users, Documents, And Systems. In Handbook of New Media, Eds. L. Lievrouw and S. Livingston, London: Sage, UK, , MILGRAM, P. AND KISHINO, F A Taxonomy of mixed reality visual displays. IEICE Transactions on Information Systems, Vol E77-D, No.12 December MILGRAM, P. AND COLQUHOUN, H. 1999a. A framework for relating head mounted displays to mixed reality displays. Proceedings of the human factors and ergonomics society, 43rd annual meeting, San Jose, MILGRAM, P. AND COLQUHOUN, H. 1999b. A Taxonomy of Real and Virtual World Display Integration, Mixed Reality - Merging Real and Virtual Worlds. In Proceedings of the first International Symposium on Mixed Reality (ISMR), Yokohama, Japan, March 1999, MILIUS, J., MacDonald, W. J. and Heller, B Rome. HD Vision Studios, HBO, Internet Site: MITCHELL, W. L Moving the museum onto the Internet: The use of virtual environments in education about ancient Egypt. In Virtual Worlds on the Internet, J.A. VINCE AND R.A. EARNSHAW, IEEE Computer Society Press, NATIONAL MUSEUM OF SCIENCE AND TECHNOLOGY National Museum of Science and Technology, Milano, Italy, Internet Site: OHTA, Y Mixed Reality: Merging Real and Virtual Worlds. Springer, ISBN: PAIANO, R Photographic Virtual Reality and Architectural Heritage. In Proceedings of the International Conference on Hypermedia and Interactivity in Museums (ICHIM), Vol.2, Milan, Italy, September 2001,

272 Appendix 1: ACM JOCCH Paper 271 PAPAGIANNAKIS, G., L'HOSTE, G., FONI, A., MAGNENAT-THALMANN, N Real-Time Photo Realistic Simulation of Complex Heritage Edifices. In Proceedings of the Seventh International Conference on Virtual Systems and Multimedia (VSMM), Berkley, USA, October 2001, IEEE Computer Society, Washington, DC, USA, PAPAGIANNAKIS, G., FONI, A., MAGNENAT-THALMANN, N. 2005a. Practical Precomputed Radiance Transfer for Mixed Reality. In Proceedings of the Virtual Systems and Multimedia Conference (VSMM), Ghent, Belgium, August 2005, VSMM Society, PAPAGIANNAKIS, G., SCHERTENLEIB, S., O'KENNEDY, B., ARÉVALO- POIZAT, M., MAGNENAT-THALMANN, N., STODDART, A. AND THALMANN, D. 2005b. Mixing Virtual and Real scenes in the site of ancient Pompeii. Computer Animation and Virtual Worlds Journal, Vol. 16, No. 1, February 2005, PAQUET, E., EL-HAKIM, S., BERALDIN, J.-A., AND PETERS, S The Virtual Museum: virtualisation of real historical environments and artifacts and three-dimensional shape-based searching. In Proceedings of the International Symposium on Virtual and Augmented Architectures (VAA), Dublin, Ireland.June 2001, PARK, C., KIM, I., AHN, S. C., KO, H., KWON, Y. AND KIM, H The design and implementation of Kyongju VR Theatre. In Proceedings of the Third International Conference on Virtual Reality and Its Application in Industry, SPIE, Volume 4756, Hangzhou, China, April 2002, International Society for Optical Engineering, Ed. Zhigeng Pan, Jiaoying Shi, RAFAELI, S Interactivity: From new media to communication. In Sage Annual Review of Communication Research, Advancing Communication Science: Merging Mass and Interpersonal Processes, R. P. HAWKINS, J. M. WIEMAN, AND S. PINGREE, Newbury Park, CA, Sage Publications, RAFAELI, S. AND ARIEL, Y Assessing Interactivity in Computer- Mediated Research. In The Oxford Handbook of Internet Psychology, JOINSON, A.N., MCKENNA, K.Y.A., POSTMES, T. AND RIEPS, U.D., Oxford University Press, Chapter 6, REYNOLDS, P. J The Nature of Experiment in Archaeology. In Experiment & Design: archaeological studies in honor of John Coles, A.F. HARDING, Oxbow Books,

273 272 Methodological approaches for the protection of cultural heritage in the digital age SCOTT, R Gladiator. DreamWorks SKG, Universal Pictures, Internet Site: SCOTT, R Kingdom of Heaven. Twentieth Century-Fox Film Corporation, Internet Site: SCHAPER, E Fiction and the Suspension of Disbelief. British Journal of Aesthetics 18, SINCLAIR, P. A. S., GOODALL, S., LEWIS, P. H., MARTINEZ, K. AND ADDIS, M. J Concept browsing for multimedia retrieval in the SCULPTEUR project. In Proceedings of the second Annual European Semantic Web Conference, Heraklion, Crete, May SORENSEN, T Project: Thoroughbred of the Sea: The trial voyage to Dublin. Research Plan, The Viking Museum, Roskilde, Denmark, electronic version available at: v2006_uk.pdf SORIN A Proposed Curriculum for Digital Heritage Studies Policies, Practices and Developments in Europe - Volume 3. EPOCH Deliverable D.4.13 Final State of the Union (2008), pp SQUIRE, K. AND BARAB, S Replaying history: engaging urban underserved students in learning world history through computer simulation games. In Proceedings of the sixth international conference on Learning sciences, Santa Monica, California, SRINIVASAN, R. AND HUANG, J Fluid ontologies for digital museums. International Journal on Digital Libraries, Issue Volume 5, Number 3 / May, 2005, STEUER, J Defining Virtual Reality: Dimensions Determining Telepresence. Journal of Communication, 42 (4,)

274 Appendix 1: ACM JOCCH Paper 273 STRICKER, D., DÄHNE, P., SEIBERT, F., CHRISTOU, I. T., ALMEIDA, L. AND IOANNIDIS, N Design and Development Issues for ARCHEOGUIDE: An Augmented Reality based Cultural Heritage On-site Guide. In Proceedings of the Euro Image ICAV 3D Conference in Augmented Virtual Environments and Three-dimensional Imaging, Mykonos, Greece, May SVANAES, D Understanding Interactivity: Steps to a Phenomenology of Human-Computer Interaction. NTNU, Trondheim, Norway, available at: THE EDUCATION ARCADE, MIT 2004.Revolution, Project Internet Site: TSICHRITZIS, D. AND GIBBS, S Virtual Museums and Virtual Realities. In Proceedings of the International Conference on Hypermedia and Interactivity in Museums, Pittsburgh, PA, USA, Archives & Museum Informatics, October 1991, UNITED KINGDOM NATIONAL ARCHIVES United Kingdom National Archives, Internet Site: earch4 UNREAL Epic Games, Inc Unreal Engine, Internet Site: VALTOLINA, S., MAZZOLENI, P., FRANZONI, S. AND BERTINO, E A semantic approach to build personalized interfaces in the cultural heritage domain. In Proceedings of the working conference on Advanced visual interfaces, Venezia, Italy, VERGAUWEN, M., WILLEMS, G., VERBIEST, F., VAN GOOL, L.J., PLETINCKX, D Combining image and model based rendering of an archaeological site. SP: IC (20), No. 9-10, October-November 2005, VIRTUAL MUSEUM OF CANADA 2008.Virtual Museum of Canada, Internet Site:

275 274 Methodological approaches for the protection of cultural heritage in the digital age VIRTUAL MUSEUM OF JAPANESE ARTS 2008.Virtual Museum of Japanese Arts, Internet Site: VISION ROMA S.R.L Past and present collection. Dian Publications, Athens, Greece. VLAHAKIS, V Design and application of the LIFEPLUS Augmented Reality System for continuous, context-sensitive guided tours of indoor and outdoor cultural sites and museums. In Proceedings of the 2003 conference on Virtual reality, archeology, and cultural heritage (VAST), Brighton, UK, November 2003, VRWAY COMMUNICATIONS Panorama of the 15th century church Santa Maria degli Angioli in Lugano, Switzerland and the frescoes by Bernardino Luini, Project Internet Site: VRMAG, WARD, A The Movie "Gladiator" in Historical Perspective. New England Classical Newsletter, William Wyatt, Jr., May 2001, electronic version available at: WOJCIECHOWSKI, R., WALCZAK, K., WHITE, M. AND CELLARY, W Building Virtual and Augmented Reality museum exhibitions. In Proceedings of the ninth international conference on 3D Web technology, Monterey, California, USA, April 2004, WRIGHT, M. W., WATSON, G., DUNLOP, G. AND MIDDLETON, R Edinburgh: 200 years of heritage through image-based virtual Environments. In Proceedings of the Fifth International Conference on Virtual Systems and Multimedia (VSMM), Dundee, UK, September 1999, VSMM Society, ZHUKOV, S., IONES, A., KRONIN, G Using Light Maps to create realistic lighting in real-time applications. In Proceedings of The Sixth International Conference in Central Europe on Computer Graphics and Visualization (WSCG), University of West Bohemia, Campus Bory, Plzen - Bory, Czech Republic, February 1998,

276 Appendix 2: Questionnaire 275 Appendix 2: Questionnaire Section A: information related to demographic attributes. This part of the questionnaire regroups some questions that are related to some variables indirectly related to the studied problematic. Such variables are mainly dependant on the personal demographic characteristics pertaining to the interviewed subject. Please check the box that represents the appropriate answer. Q01: your age is comprised between: 15 and 20 years 21 and 25 years 26 and 30 years 31 and 35 years 36 and 40 years 41 and 45 years 46 and 50 years 51 and 55 years 56 and 60 years More than 61 years Q02: what is your gender? Male Female

277 276 Methodological approaches for the protection of cultural heritage in the digital age Section B: education This part of the questionnaire regroups some questions related to the type of schools you attended and the different degrees that you obtained. Q03: between the following degrees, which one did you obtain during your formation? (please check the box corresponding to the more recent or advanced degree you obtained if more options are applicable) Compulsory school (primary school) Maturity (secondary school) Professional Maturity or Capacity Certificate Secondary Technical Schools University Degree Professional University Degree Master Degree PhD Other Q04: in which of the following areas did you specialized? : History or History of Art Archeology Documentation Architecture Technical drawing Artistical related activities Restoration or Conservation Engineering Informatics Multimedia Music Other

278 Appendix 2: Questionnaire 277 Section C: preferences concerning virtual reconstruction strategies applicable to cultural heritage items This part of the questionnaire is related to your preferences concerning the possible visualization strategies that could be employed to achieve a visual representation of cultural heritage items. Q05: how important or useful is for you the possibility to represent in 2 dimensions (CAD) a cultural item? Not important at all Marginally important Rather important Very important Q06: how important or useful is for you the possibility to represent in 3 dimensions a cultural item? Not important at all Marginally important Rather important Very important Q07: considering the following example (the Roman theatre of Aspendos as it appears today), which option would you adopt for a physical restoration of the building? Complete reconstruction using modern materials Complete reconstruction using materials similar to the original ones Reconstruction of some of its elements using modern materials Reconstruction of some of its elements using materials similar to the original ones Interventions aimed at consolidating the built structure only, in order to better preserve the monument No intervention at all (the monument shall not be touched in any way)

279 278 Methodological approaches for the protection of cultural heritage in the digital age Q08: considering the same monument, if it would be possible to realize a virtual 3D computer model to represent such edifice, would you prefer to: Visualize it only as it is today (left) Visualize it only as it was in certain historical period (right) Visualize it as it was in several historical periods Not to proceed to any computer visualization at all Q09: how important or useful is the possibility to represent a cultural item as it appeared in the past in a given historical period? Not important at all Marginally important Rather important Very important Q10: how important or useful is the possibility to represent a cultural item exactly as it appears today? Not important at all Marginally important Rather important Very important Q11: a virtual reconstruction ought to be primarily: Beautiful, visually appealing or pleasing Precise Historically or scientifically valid Useful or functional to specific real needs (i.e. it should prove an efficient tool to transmit pertinent information more easily)

280 Appendix 2: Questionnaire 279 Q12: how important or useful is that a virtual representation of a cultural heritage item exhibits a visually appealing presentation? Not important at all Marginally important Rather important Very important Q13: how important or useful is that a virtual representation of a cultural heritage item exhibits a high degree of precision (average error inferior to 1cm)? Not important at all Marginally important Rather important Very important Q14: how important or useful is that a virtual representation of a cultural heritage item exhibits a high degree of scientific or historical validity? Not important at all Marginally important Rather important Very important Q15: considering the following possibilities, would you favour a virtual reconstruction that is: Very precise, but not particularly visually appealing Very beautiful, but not particularly precise If the reconstruction is historically valid, it does not matter if it is very precise or visually appealing

281 280 Methodological approaches for the protection of cultural heritagee in the digital age Q16: the virtual reconstruction of a cultural heritage item that missing structural elements shall include: presents No additions at all (faithful representation of its present state) Small additions (only the main known elements on which we have sufficient data to allow a scientifically valid and historically correct reconstruction of their properties) Important additions based on reasonable hypothesis (reconstruct whenever possible without resorting to speculative hypothesis to recreate missing elements on which we have no certified information) A complete visualisation of the cultural heritage item presented as it was in its original state (completely reconstructed, resorting if necessary to the implementation of speculative hypothesis and nonn certified deductions whenever insufficient data is available to determine the properties of the missing elements)

282 Appendix 2: Questionnaire 281 Q17: in the event that a complete reconstruction of missing elements, on which insufficient data is available, had to be implemented, to determine their visual properties would you prefer to: Extrapolate or build some hypothesis using as a base the remaining similar elements still existing on the historical site Use exclusively certified hypothesis and historical studies scientifically valid Use as a reference structural elements that are still existing today in similar sites of the same historical period To introduce such elements in the virtual reconstruction anyways, however they should be clearly presented and visually easily identifiable as additions (coloured in red or neutral grey for instance) Q18: how important or useful is the possibility to include in the virtual representation of cultural heritage item the visualisation of its lost or destroyed structural elements? Not important at all Marginally important Rather important Very important Q19: the virtual reconstruction of a cultural heritage items should: Only feature a general reconstruction of the shape of the edifice with the correct visualisation of its volumes and proportions Feature a reconstruction of its main elements without visualizing the small details Feature a reconstruction as detailed as possible

283 282 Methodological approaches for the protection of cultural heritage in the digital age Q20: how important or useful is the possibility to include in the virtual representation of a cultural heritage item the precise visualisation of its small details? Not important at all Marginally important Rather important Very important Q21: the virtual reconstruction of a cultural heritage item should include a realistic and coherent representation of the properties of its materials and surfaces? No, a representation employing simple colours to represent the materials is sufficient Yes, a visually realistic and coherent reconstruction is sufficient, even though the parameters used to achieve their look are not physically simulated Yes, however, beside having a realistic look, the parameters used to represent and simulate the materials shall also be physically correct I think that this is not important Q22: how important or useful is the possibility to include in the virtual representation of a cultural heritage item a realistic visualisation of the properties of its surfaces and materials? Not important at all Marginally important Rather important Very important

284 Appendix 2: Questionnaire 283 Q23: the virtual reconstruction of a cultural heritage site should exhibit a simulation of its lighting conditions that is: A simple illumination model is sufficient if the shapes and volumes are clearly distinguishable and identifiable Sufficiently realistic to allow the visualisation to be visually pleasing Based on physically correct simulations in order to exactly reproduce the real lighting conditions on the real site I think that this is not important Q24: how important or useful is the possibility to include in the virtual representation of a cultural heritage item an extremely realistic simulation of its lighting conditions? Not important at all Marginally important Rather important Very important

285 284 Methodological approaches for the protection of cultural heritage in the digital age Q25: the virtual reconstruction of a cultural heritage site should: Only visualize the reconstructed edifices without their surroundings Include a simplified reconstruction of the general features of their surrounding terrain and environment for a radius of 100m-300m Include a medium complexity reconstruction of the main features of their surrounding terrain and environment for a radius of 100m-300m Include a detailed and precise reconstruction of all the features of their surrounding terrain and environment (with trees, detailed textures for the appearance of the terrain, etc.) Include a very realistic reconstruction of all the features of their surrounding terrain and environment for a larger radius (500m-1km) including houses, additional structures, realistic trees and terrain I think that this is not important Q26: how important or useful is the possibility to include in the virtual representation of a cultural heritage item a precise visualisation of the geographic features exhibited by its surrounding environment? Not important at all Marginally important Rather important Very important

286 Appendix 2: Questionnaire 285 Q27: the virtual reconstruction of a cultural heritage site should also include the representation of modern elements (such as streets, houses or recent interventions and modifications operated on the structure of the heritage item)? No, the site should only be represented as it was in a determined past historical period Yes, the site should be visualized as it was in the past but presented in its modern context The site should be visualized only as it is today Multiple visualisations should be prepared: the site should be presented both as it was in the past and as it is today in its modern context Q28: In the eventuality that a reconstruction of complex missing elements shall be realized (such as frescoes, painted decorations, architectural elements or parts of the main structure) but the available data only allows to determine their geometry but not the properties of their materials, would you prefer: To exclude completely such elements from the visualisation To include such elements in the visualisation and visually integrating them with the known parts of the edifice assigning to them similar materials To include such elements in the visualisation, however they should be clearly presented and easily identifiable as additions (coloured in red or neutral grey for instance)

287 286 Methodological approaches for the protection of cultural heritage in the digital age Q29: In the eventuality that the present conservation state of an heritage item exhibits a clear and strong degradation of some of its decorated surfaces (due to humidity, water infiltrations or missing parts) that caused a modification of their original shape, appearance or colour, would you prefer that in the virtual reconstruction: To visualize such surfaces as they are today in their present damaged state To visualize the damaged parts of such surfaces with a simple colour (for instance using a neutral grey) To proceed to a virtual restoration of the damaged surfaces to recover their original appearance To completely exclude such surfaces form the virtual representation Q30: The virtual representation of a cultural heritage site shall also include the visualisation of virtual humans? Yes, however the virtual humans should only represent contemporary modern people Yes, however the virtual humans should only represent people belonging to the same context and historical period exhibited by the presented reconstruction of the heritage site Two models should be realized: with and without virtual humans No, the inclusion of virtual humans is not necessary I think that this is not important

288 Appendix 2: Questionnaire 287 Q31: In the eventuality that some virtual humans had to be included in the virtual representation of a cultural heritage site, would you favour: (maximum 2 answers): Static virtual humans (they do not move nor perform any action) Virtual humans that perform unspecified general actions (such as walking or speaking to one another) Virtual humans that perform specific actions associated with plausible activities that could have been performed at the site in the historical period presented by the virtual reconstruction (the virtual humans behave according to a specific scenario) Virtual humans that perform any action determined interactively by the user Virtual humans that react to the presence of the user providing information on the heritage site and the historical period of the reconstruction, and that answer to the questions of the user Q32: how important or useful is the possibility to include in the virtual visualisation of a cultural heritage item the representation of virtual humans? Not important at all Marginally important Rather important Very important Q33: which kind of virtual reconstruction would you favour between the following possibilities? A virtual reconstruction visually engaging, extremely photo realistic but not interactive (such as a movie) A virtual reconstruction visually not very realistic, but extremely interactive (such as a videogame where is possible to freely interact with the virtual humans) A virtual reconstruction visually pleasing featuring a limited interactivity (the virtual humans are following scripted predetermined scenarios) I think that this is not important

289 288 Methodological approaches for the protection of cultural heritage in the digital age Q34: how important or useful is the possibility to visualize interactively the virtual representation of a cultural heritage (such as freely changing the point of view or freely exploring the virtual model)? Not important at all Marginally important Rather important Very important Q35: in the following list of activities connected to cultural heritage protection, which ones do you think could benefit the most from the contribution that virtual reconstructions could offer? For which ones the use of virtual reconstructions could represent a valid asset? Documentation of cultural heritage Analysis, study or investigation of the cultural heritage Conservation or restoration of the cultural heritage Defence or protection of the cultural heritage Dissemination of information connected to the cultural heritage None of the above fields Q36: for which of the following uses do you think virtual reconstructions of cultural heritage items are best suited? (Maximum 3 answers) For entertainment For education For dissemination and divulgation (transmit more easily, or faster, pertinent information on the cultural item) To allow more efficient approaches (or to increase the rapidity) To offer new possibilities or new approaches (increase the quality) For research, simulation or scientific experimentation For the documentation of the cultural heritage To preserve and archive more effectively all the information relative to cultural heritage items To provide an intuitive way to access additional pertinent information linked to the cultural heritage item (to employ the reconstruction as an interface) To optimise and better plan the efforts associated with an intervention on the heritage item (pre-visualize the areas of intervention) Employing a virtual model could certainly constitute an asset, however I think that its real contribution is not decisive I think that employing a virtual model does cannot provide any advantage nor might constitute a valid asset

290 Appendix 2: Questionnaire 289 Q37: the final use that you would do of a virtual representation of a cultural heritage item could ideally be compared to: (maximum 2 answers) An entertaining interactive experience (such as a videogame) A passive entertaining experience (such as watching a movie) Watching fixed images or photographic pictures A way to access more easily and rapidly pertinent information linked to the virtual model of the cultural item (such as clicking on a painting or an altar to obtain and be presented with detailed information on its characteristics and properties) A way to access cultural information in a more easy and accessible fashion An educative tool A professional working tool Q38: having at your disposal a virtual representation of an heritage site, would you prefer: (maximum 2 answers) To sit in front of a screen and watch it as a movie or a documentary Observe some virtual humans dressed according to the presented historical period perform a scenario based on plausible historical data Interact with the simulation and interactively change the position from which you observe the details of the reconstruction (freely explore and visualize the model) To go on the real site and, with some special equipment such as special glasses, visualize the reconstructed virtual elements or humans superimposed on the real still existing parts of the structures of the site To freely interact with the virtual humans dressed as historical characters in the simulation (ask them questions and receive pertinent answers) Q39: how much time are you willing to invest, or wait, for the preparation of a virtual model of a cultural heritage item that fully meets your main practical needs, professional requirements and general expectations as end user? A few hours A few days A few weeks A few months If the final result meets my needs and requirements, I would wait as long as it were necessary to prepare such a model

291 290 Methodological approaches for the protection of cultural heritage in the digital age Section D: preferences concerning visualisation techniques and methods applicable to display cultural heritage virtual reconstructions This section of the questionnaire regroups some questions related to your preferences concerning the implementation of some techniques and visualisation technologies that can be used to virtually represent s cultural heritage item. Please check the box that corresponds to the image that you prefer or that you feel the most appropriate or useful in relation to your specific professional needs and requirements. Q40: between manual methods and CAD methods, you favour: Restitution drawings realized by hand (above left) Plans and sections realized by hand (above right) Plans and sections realized with CAD tools (below left) Reconstructions and visualisations realized with CAD tools (below right)

292 Appendix 2: Questionnaire 291 Q41: between 2D representations and 3D representations, you favour: Photographs (left) 2D plans and sections realized with CAD tools (center) 3D models and representations realized with CAD tools (right) Q42: between representations featuring the visualisation of the surrounding environment and models not exhibiting any geographical context, you favour: Representations without any visualisation of the environment (left) Representations featuring a simplified visualisation of their environment (center) Representations featuring a detailed visualisation of their environment (right)

293 292 Methodological approaches for the protection of cultural heritage in the digital age Q43: between static images or movies and interactive models, you favour: Static images or movies (above left) Simple interactive models (above right) Realistic interactive models (below) Q44: between detailed or precise models and simplified or generic models exhibiting only the main volumes and shapes, you favour: Very precise and detailed models (left) Models only featuring the representation of their general shapes and volumes (right)

294 Appendix 2: Questionnaire 293 Q45: between models featuring the inclusion of virtual humans and models not exhibiting their inclusion, you favour: Models without virtual humans (left) Models with the inclusion of virtual humans (right) Q46: between representations featuring a simple illumination model and representations exhibiting a complex simulation of their lighting conditions, you favour: Representation featuring simple illumination models (left) Representation featuring complex illumination models (right)

295 294 Methodological approaches for the protection of cultural heritage in the digital age Q47: between Representations exhibiting a high degree of visual realism (visually not geometrically, the proposed representations present in fact the same amount of detail and complexity) and representations featuring a lesser degree of visual realism, you favour: Representations exhibiting a very simple visual realism (above left) Representations exhibiting a medium degree of visual realism (above right) Representation exhibiting a high degree of visual realism (below) Q48: between representations featuring the inclusion of real elements and representations not exhibiting such inclusion, you favour: Representations featuring only real elements, such as photographs (left) Representations featuring real and virtual elements superimposed (right) Representations featuring only virtual elements