CS415 Human Computer Interaction Lecture 10 Advanced HCI Universal Design & Intro to Cognitive Models October 30, 2017 Sam Siewert
Summary of Thoughts on Intelligent Transportation Systems Collective Wisdom of Our Classes (2015, 2016, 2017) Self driving Cars will become ubiquitous by? Year 2020 2025 2030 2035 2040 2045 Beyond Out of 2015 class 1 1 1 0 1 0 0 6 2016 class 0 1 3 0 1 0 0 5 2017 class 0 0 1 1 2 1 2 7 Sam Siewert 2
Summary of Thoughts on ITS Revolutionary vs. Evolutionary from Now to Ubiquitous Evolutionary [7] Self-driving cars have to deal with personal acceptance, like the touch screens did, but these cars also have to deal with safety By 2035 the (sic) millenial generation will have a big push on automation and technology the Baby Boomer generation will be getting to a point where they are out numbered. evolution of self-driving cars will be slow and steady with many government and safety hoops like manual transmission vehicles, modern vehicles (human driven) will stick around It will never become a nationwide reality most people enjoy driving a small percentage of the population will benefit from the technology like people with disabilities evolution with each generation going more towards self driving contingent upon whether or not self-driving does not have any large number of (sic) causulties or negative impact as the generations shift self-driving will be the last step of this evolutionary process each dealer will eventually come up with their own design enough time to develop software and hardware to be deemed safe enough time to be deemed socially acceptable younger generations are more comfortable with the thought of self driving The key to this problem is social acceptance Revolutionary [0] - No responses interpreted as revolutionary Sam Siewert 3
Assignments Assignment #4 Posted, Last Practice Lab (Cameras and OpenCV) Assignment #5 Propose Group Project (Groups of 3) Assignment #6 Project Final Oral Exam Presentation of Project HCI Design Experimental Design for Usability Analysis Prototype or Mockup of Design Sam Siewert 4
This Week Chapter 10 Introduction to Universal Design Making Computers Available to Everyone Computer Solutions for Prosthetics Active Learning Discussion The Human Cyborg? Chapter 12 Cognitive Models GOMS Goal, Operators, Methods and Selection Goal - describing what the user wants to achieve Operators - basic actions that the user must perform in order to use the system Methods - several ways in which a goal can be split into subgoals Selection choice of method and prediction of which will be used Sam Siewert 5
Assignment #4 Use of Digital Cameras with Embedded Linux (V4L2/UVC/USB) - OpenCV Applications in Intelligent Transportation, Augmented Reality, Communications (e.g. Skype, Apple Facetime, WeChat) Jetson has OpenCV installed (JetPack 3.1) Sam Siewert 6
Assignment #4 - Objectives Play with the Cameras and OpenCV First Verify Hardware Get a Feel for Performance Capabilities Learn about Camera Devices and Device Interfaces Develop a Camera Application Consider Processing for Display and Interaction Human Aspects Color, Frame Rate, Segmentation Sam Siewert 7
Assignment #4 Demo Real-time Camera Scene Analysis Uncompressed, Un-encoded Camera Frames (Not MPEG) Transforms RGB or Graymap XY Pixel Array Into Segments and Major Components in Frames Eliminate Backround (Difference of Frames Current, Historical) Segments Analyzed (Recognition) e.g. Gesture (Thumbs Up, Thumbs Down) Segmenting Linear Elements Segmenting Circular Elements Sam Siewert 8
Universal Design and Embedded HCI Chapter 10 Introduction to Universal Design Making Computers Available to Everyone Computer Solutions for Prosthetics Active Learning Discussion The Human Cyborg? Popular News Ted Talk on Color Blindness Ted Talk We are Already Cyborgs Multi-modal Systems Speech Audio Touch Handwriting Gestures Sam Siewert 9
Humans have long been using tools to extend or restore physical capability, and instruments to enhance perception of the world, which fits the definition of cyborg (cyber organism). Minute Paper https://en.wiktionary.org/wiki/cyborg Tools and instruments which progress to wearable devices, which can in turn become prosthetics, are a form of personal cybernetics, but can also be used for social networking as well as personal perception. Is this a positive and useful form of HCI? What costs, benefits, and risks do you see related? Sam Siewert 10
Design for Diversity Universal Design People with Sensory, Physical or Cognitive Impairment Age Groups Varying Cultures and Backgrounds Localization Languages 7 Principles of Universal Design 1. Equitable Use No User is Excluded 2. Flexibility Methods of Use, Pace, Precision, Ability 3. Intuitive Wide Range of Knowledge and Experience 4. Perceptible Information Graphic, Verbal, Text, Touch 5. Error Tolerance Minimize Impact of Mistakes (e.g. Undo) 6. Low Physical Effort Comfortable, Minimal Fatigue 7. Size and Space Sitting or Standing, Body Size, Posture, Mobility Sam Siewert 11
Speech Generation Text to Speech - Synthesis Recognition Speech to Text, to Syntax and Semantics Sam Siewert 12
Speech Recognition Better Today, But Still in Limited Use Google Demonstration Page Sam Siewert 13
Speech Synthesis Screen Readers, Web Readers http://www.bobby-approved.com/ Sam Siewert 14
Uses for Speech Recognition and Synthesis Where and When to Use Sam Siewert 15
Haptic Interfaces Haptics - Touch Cutaneous (skin tactile sensations) Kinesthetics movement perception and position Current Use Surgical Training, e.g. LapSim Sam Siewert 16
Handwriting Systems Record e.g. Signatures (FedEx, UPS, DocuSign, ) AOS Swipe in Settings (Hybrid Handwriting/Typing) Recognize Input No Systems Provide General Cursive Script Recognition Some Success with Character at a Time Input Apple Newton (Limited Success) Touch Screen Typing Apple Newton Stylus PDA Sam Siewert 17
Gesture Recognition Cameras (Active or Passive) with Depth Mapping, Segmentation and Tracking Data Gloves Rotation (Yaw, Pitch, Roll), Translation (X,Y,Z), Rates of Translation/Rotation, and Finger Flexures Camera Alone, Glove Alone, or Sensor Fusion Glove is Intrusive, but Improves Accuracy Challenge Full Speed American Sign Language Recognition Sam Siewert 18
3D Active Computational Photometry Concept (Rev-A + TI Kit) TI DLP Light-crafter Kit http://www.ti.com/tool/dlplightcrafter IR Pattern Projection Photo credits and reference: Dr. Daniel Aliaga, Purdue University https://www.cs.purdue.edu/homes/aliaga/ Analog Camera #2 (Near Infrared) Analog Camera #1 RGB (Visible) Altera FPGA CVPU (Computer Vision Processing Unit) Depth Map HD Digital Camera Port (Snapshot) USB 2.0, PCIe Host Channels Flash SD Card Networked Video Analytics Mobile Sensor Network Processor (TI OMAP, Atom) https://www.cs.purdue.edu/homes/aliaga/cs635-10/lec-structured-light.pdf Sam Siewert 19
3D Computer Vision Transforms Long Range ( > 5 meters) Using Passive Binocular Methods Impractical to Project from a UAV or Long Range Observer Requires Image Registration Accurate Camera Intrinsic (Camera Characteristics) & Extrinsic (e.g. Baseline) Short Range ( < 5 meters), Structured IR Light Projection for RGB-D Compare to ASUS Xtion and PrimeSense Off-the-Shelf Robust Depth Maps with Less Noise Showing Significant Promise to Improve CV Scene Segmentation and Object Recognition Compared to 2D Change Their Perception, By Xiaofeng Ren, Dieter Fox, and Kurt Konolige, IEEE RAS, December 2013. Noise in Passive Depth Maps Change Their Perception, By Xiaofeng Ren, Dieter Fox, and Kurt Konolige, IEEE RAS, Sam Siewert Robust Active Depth Map December 2013. 20
Design for Users with Disabilities Minority of Market, but Support Required Visual Impairment Hearing Impairment Physical Impairment Speech Impairment Cognitive Impairments Dyslexia, Autisim Sam Siewert 21
Cultural and Age Considerations Elderly, Age 50+ Tablet Children, Kids Tablet e.g. IPAD Mini 2, Amazon Fire Kids Edition, etc. Language Localization Fundamental Translation, Keyboard, Graphics for Native Language Support Cultural Differences E.g. Nodding or Shaking Head in Agreement Global Marketplace High Value Today with Globalization Sam Siewert 22
Cognitive Models - Introduction Chapter 11 of HCI - User Support Systems (Built-in Help) Chapter 12 of HCI - Cognitive Models Hierarchical Models Task and Goal Structure GOMS (Goals, Operators, Methods, Selection), p. 421-425 Goal - what you want to achieve Operators - basic actions Methods - splitting goal into many sub-goals Selection - choice of method for more than one Linguistic Models System Grammar Physical and Device Models Human Motor Skills Cognitive Models Combine All 3 of the Above Sam Siewert 23