Supplementary Material for Generative Adversarial Perturbations
|
|
|
- Rosalyn Welch
- 5 years ago
- Views:
Transcription
1 Supplementary Material for Generative Adversarial Perturbations Omid Poursaeed 1,2 Isay Katsman 1 Bicheng Gao 3,1 Serge Belongie 1,2 1 Cornell University 2 Cornell Tech 3 Shanghai Jiao Tong University {op63,isk22,bg455,sjb344}@cornell.edu 1. Runtime Analysis Note that inference time is not an issue for universal perturbations as we just need to add the perturbation to the input image during inference. Therefore, we provide running time only for image-dependent perturbations. In this case, we need to forward the input image to the generator and get the resulting perturbation. Table 1 demonstrates the inference time for image-dependent perturbations. It also shows the generator s architecture for each task including the number of filters in the first layer. We perform model-level parallelization across two GPUs, and batch size is set to be one. Notice that inference time is in the order of milliseconds, allowing us to generate perturbations in real-time. Table 2 shows inference time for the segmentation task. Two architectures with similar performance are given. Here we deal with images in the Cityscapes dataset, and we need models with more capacity; hence, the inference time is larger compared with the classification task. Task Architecture Titan Xp Tesla K40 Non-targeted ResNet Gen. 6 blocks, 0.27 ms 4.7 ms 50 filters Targeted ResNet Gen. 6 blocks, 57 filters 0.28 ms 4.8 ms Table 1: Average inference time per image and generator s architecture for image-dependent classification tasks. Target model is Inception-v3. 2. Resistance to Gaussian Blur We examine the effect of applying Gaussian filters to perturbed images. Results for the classification task are shown in Table 3. In order to be comparable with [26], we consider non-targeted image-dependent perturbations with Destruction Rate (fraction of images that are no longer misclassified after blur) as the metric. For most σ values, our method is more resistant to Gaussian blur than I-FGSM. Architecture Titan Xp Tesla K40m U-Net Generator: 8 layers, 200 filters ms ms ResNet Generator: 9 blocks, 145 filters ms ms Table 2: Average inference time per image and generator s architecture for the semantic segmentation task. Targeted image-dependent perturbations are considered with FCN-8s as the pre-trained model. We also evaluate the effect of Gaussian filters for the segmentation task. Results are given in Table 4. As we can observe, the perturbations are reasonably robust to Gaussian blur. σ = 0.5 σ = 0.75 σ = 1 σ = 1.25 GAP 0.0% 0.8% 3.2% 8.0% I-FGSM 0.0% 0.5% 8.0% 23.0% Table 3: Destruction Rate of non-targeted image-dependent perturbations for the classification task. Perturbation norm is set to L = 16. σ = 0.5 σ = 0.75 σ = 1 σ = 1.25 L = % 76.9% 66.0% 57.1 % L = % 90.1% 80.0% 69.6% L = % 95.7% 89.3% 78.8% Table 4: Success rate of targeted image-dependent perturbations for the segmentation task after applying Gaussian filters. 3. Additional Examples More examples of both classification and segmentation adversarial perturbations are given in the following figures. 1
2 (a) Target model: VGG-19, Fooling ratio: 94.9% (b) Target model: VGG-16, Fooling ratio: 93.9% Figure 1: Non-targeted universal perturbations. From top to bottom: original image, enhanced perturbation and perturbed image. Perturbation norm is set to L 2 = 2000 for (a) and (b) and to L = 10 for (c) and (d).
3 (c) Target model: Inception-v3, Fooling ratio: 79.2% (d) Target model: VGG-19, Fooling ratio: 80.1% Figure 1: Non-targeted universal perturbations (continued). From top to bottom: original image, enhanced perturbation and perturbed image. Perturbation norm is set to L 2 = 2000 for (a) and (b) and to L = 10 for (c) and (d).
4 (a) Target: Jigsaw Puzzle, Top-1 target accuracy: 89.3% (b) Target: Teapot, Top-1 target accuracy: 62.2% Figure 2: Targeted universal perturbations. From top to bottom: original image, enhanced perturbation and perturbed image. Perturbation norm is set to L = 10, and target model is Inception-v3.
5 (c) Target: Chain, Top-1 target accuracy: 64.9% (d) Target: Hamster, Top-1 target accuracy: 60.0% Figure 2: Targeted universal perturbations (continued). From top to bottom: original image, enhanced perturbation and perturbed image. Perturbation norm is set to L = 10, and target model is Inception-v3.
6 (a) L = 7 (b) L = 10 (c) L = 13 Figure 3: Non-targeted image-dependent perturbations. From top to bottom: original image, enhanced perturbation and perturbed image. Three different thresholds are considered with Inception-v3 as the target model.
7 (a) Target: Jigsaw puzzle, Top-1 target accuracy: 98.1% (b) Target: Knot, Top-1 target accuracy: 95.0% Figure 4: Targeted image-dependent perturbations. From top to bottom: original image, enhanced perturbation and perturbed image. Perturbation norm is set to L = 10, and Inception-v3 is the pre-trained model.
8 (c) Target: Chain, Top-1 target accuracy: 89.7% (d) Target: Teapot, Top-1 target accuracy: 90.6% Figure 4: Targeted image-dependent perturbations (continued). From top to bottom: original image, enhanced perturbation and perturbed image. Perturbation norm is set to L = 10, and Inception-v3 is the pre-trained model.
9 (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Target (f) Prediction for perturbed image Figure 5: Targeted universal perturbations with L = 5. Zoom in for details. (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Target (f) Prediction for perturbed image Figure 6: Targeted universal perturbations with L = 10.
10 (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Target (f) Prediction for perturbed image Figure 7: Targeted universal perturbations with L = 20. (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Target (f) Prediction for perturbed image Figure 8: Targeted image-dependent perturbations with L = 5. Zoom in for details.
11 (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Target (f) Prediction for perturbed image Figure 9: Targeted image-dependent perturbations with L = 10. (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Target (f) Prediction for perturbed image Figure 10: Targeted image-dependent perturbations with L = 20.
12 (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Groundtruth (f) Prediction for perturbed image Figure 11: Non-targeted universal perturbations with L = 5. Zoom in for details. (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Groundtruth (f) Prediction for perturbed image Figure 12: Non-targeted universal perturbations with L = 10.
13 (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Groundtruth (f) Prediction for perturbed image Figure 13: Non-targeted universal perturbations with L = 20. (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Groundtruth (f) Prediction for perturbed image Figure 14: Non-targeted image-dependent perturbations with L = 5. Zoom in for details.
14 (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Groundtruth (f) Prediction for perturbed image Figure 15: Non-targeted image-dependent perturbations with L = 10. (a) Original image (b) Perturbation (c) Perturbed image (d) Prediction for original image (e) Groundtruth (f) Prediction for perturbed image Figure 16: Non-targeted image-dependent perturbations with L = 20.
Convolutional Neural Networks
Convolutional Neural Networks Convolution, LeNet, AlexNet, VGGNet, GoogleNet, Resnet, DenseNet, CAM, Deconvolution Sept 17, 2018 Aaditya Prakash Convolution Convolution Demo Convolution Convolution in
Adversarial Attacks on Face Detectors using Neural Net based Constrained Optimization
Adversarial Attacks on Face Detectors using Neural Net based Constrained Optimization Joey Bose University of Toronto joey.bose@mail.utoronto.ca September 26, 2018 Joey Bose (UofT) GeekPwn Las Vegas September
Semantic Segmentation on Resource Constrained Devices
Semantic Segmentation on Resource Constrained Devices Sachin Mehta University of Washington, Seattle In collaboration with Mohammad Rastegari, Anat Caspi, Linda Shapiro, and Hannaneh Hajishirzi Project
Convolutional Networks for Image Segmentation: U-Net 1, DeconvNet 2, and SegNet 3
Convolutional Networks for Image Segmentation: U-Net 1, DeconvNet 2, and SegNet 3 1 Olaf Ronneberger, Philipp Fischer, Thomas Brox (Freiburg, Germany) 2 Hyeonwoo Noh, Seunghoon Hong, Bohyung Han (POSTECH,
NU-Net: Deep Residual Wide Field of View Convolutional Neural Network for Semantic Segmentation
NU-Net: Deep Residual Wide Field of View Convolutional Neural Network for Semantic Segmentation Mohamed Samy 1 Karim Amer 1 Kareem Eissa Mahmoud Shaker Mohamed ElHelw Center for Informatics Science Nile
Lecture 11-1 CNN introduction. Sung Kim
Lecture 11-1 CNN introduction Sung Kim 'The only limit is your imagination' http://itchyi.squarespace.com/thelatest/2012/5/17/the-only-limit-is-your-imagination.html Lecture 7: Convolutional
arxiv: v2 [cs.lg] 13 Oct 2018
![arxiv: v2 [cs.lg] 13 Oct 2018](/thumbs/89/100150009.jpg "arxiv: v2 [cs.lg] 13 Oct 2018")
A Systematic Comparison of Deep Learning Architectures in an Autonomous Vehicle Michael Teti 1, William Edward Hahn 1, Shawn Martin 2, Christopher Teti 3, and Elan Barenholtz 1 arxiv:1803.09386v2 [cs.lg]
Deformable Convolutional Networks
Deformable Convolutional Networks Jifeng Dai^ With Haozhi Qi*^, Yuwen Xiong*^, Yi Li*^, Guodong Zhang*^, Han Hu, Yichen Wei Visual Computing Group Microsoft Research Asia (* interns at MSRA, ^ equal contribution)
DEEP LEARNING ON RF DATA. Adam Thompson Senior Solutions Architect March 29, 2018
DEEP LEARNING ON RF DATA Adam Thompson Senior Solutions Architect March 29, 2018 Background Information Signal Processing and Deep Learning Radio Frequency Data Nuances AGENDA Complex Domain Representations
Colorful Image Colorizations Supplementary Material
Colorful Image Colorizations Supplementary Material Richard Zhang, Phillip Isola, Alexei A. Efros {rich.zhang, isola, efros}@eecs.berkeley.edu University of California, Berkeley 1 Overview This document
DSNet: An Efficient CNN for Road Scene Segmentation
DSNet: An Efficient CNN for Road Scene Segmentation Ping-Rong Chen 1 Hsueh-Ming Hang 1 1 National Chiao Tung University {james50120.ee05g, hmhang}@nctu.edu.tw Sheng-Wei Chan 2 Jing-Jhih Lin 2 2 Industrial
Supplementary Materials for
advances.sciencemag.org/cgi/content/full/1/11/e1501057/dc1 Supplementary Materials for Earthquake detection through computationally efficient similarity search The PDF file includes: Clara E. Yoon, Ossian
Deep Learning for Infrastructure Assessment in Africa using Remote Sensing Data
Deep Learning for Infrastructure Assessment in Africa using Remote Sensing Data Pascaline Dupas Department of Economics, Stanford University Data for Development Initiative @ Stanford Center on Global
Adversarial Robustness for Aligned AI
Adversarial Robustness for Aligned AI Ian Goodfellow, Staff Research NIPS 2017 Workshop on Aligned Artificial Intelligence Many thanks to Catherine Olsson for feedback on drafts The Alignment Problem (This
Road detection with EOSResUNet and post vectorizing algorithm
Road detection with EOSResUNet and post vectorizing algorithm Oleksandr Filin alexandr.filin@eosda.com Anton Zapara anton.zapara@eosda.com Serhii Panchenko sergey.panchenko@eosda.com Abstract Object recognition
A Fuller Understanding of Fully Convolutional Networks. Evan Shelhamer* Jonathan Long* Trevor Darrell UC Berkeley in CVPR'15, PAMI'16
A Fuller Understanding of Fully Convolutional Networks Evan Shelhamer* Jonathan Long* Trevor Darrell UC Berkeley in CVPR'15, PAMI'16 1 pixels in, pixels out colorization Zhang et al.2016 monocular depth
Analysis of adversarial attacks against CNN-based image forgery detectors
Analysis of adversarial attacks against CNN-based image forgery detectors Diego Gragnaniello, Francesco Marra, Giovanni Poggi, Luisa Verdoliva Department of Electrical Engineering and Information Technology
Detection and Segmentation. Fei-Fei Li & Justin Johnson & Serena Yeung. Lecture 11 -
Lecture 11: Detection and Segmentation Lecture 11-1 May 10, 2017 Administrative Midterms being graded Please don t discuss midterms until next week - some students not yet taken A2 being graded Project
Adversarial examples in Deep Neural Networks. Luiz Gustavo Hafemann Le Thanh Nguyen-Meidine
Adversarial examples in Deep Neural Networks Luiz Gustavo Hafemann Le Thanh Nguyen-Meidine Agenda Introduction Attacks and Defenses NIPS 2017 adversarial attacks competition Demo Discussion 2 Introduction
Automatic tumor segmentation in breast ultrasound images using a dilated fully convolutional network combined with an active contour model
Automatic tumor segmentation in breast ultrasound images using a dilated fully convolutional network combined with an active contour model Yuzhou Hu Departmentof Electronic Engineering, Fudan University,
arxiv: v1 [stat.ml] 10 Nov 2017
![arxiv: v1 [stat.ml] 10 Nov 2017](/thumbs/95/123582764.jpg "arxiv: v1 [stat.ml] 10 Nov 2017")
Poverty Prediction with Public Landsat 7 Satellite Imagery and Machine Learning arxiv:1711.03654v1 [stat.ml] 10 Nov 2017 Anthony Perez Department of Computer Science Stanford, CA 94305 aperez8@stanford.edu
Biologically Inspired Computation
Biologically Inspired Computation Deep Learning & Convolutional Neural Networks Joe Marino biologically inspired computation biological intelligence flexible capable of detecting/ executing/reasoning about
On the Robustness of Deep Neural Networks
On the Robustness of Deep Neural Networks Manuel Günther, Andras Rozsa, and Terrance E. Boult Vision and Security Technology Lab, University of Colorado Colorado Springs {mgunther,arozsa,tboult}@vast.uccs.edu
Deep Learning. Dr. Johan Hagelbäck.
Deep Learning Dr. Johan Hagelbäck johan.hagelback@lnu.se http://aiguy.org Image Classification Image classification can be a difficult task Some of the challenges we have to face are: Viewpoint variation:
Transformation to Artificial Intelligence with MATLAB Roy Lurie, PhD Vice President of Engineering MATLAB Products
Transformation to Artificial Intelligence with MATLAB Roy Lurie, PhD Vice President of Engineering MATLAB Products 2018 The MathWorks, Inc. 1 A brief history of the automobile First Commercial Gas Car
Fully Convolutional Networks for Semantic Segmentation
Fully Convolutional Networks for Semantic Segmentation Jonathan Long* Evan Shelhamer* Trevor Darrell UC Berkeley Presented by: Gordon Christie 1 Overview Reinterpret standard classification convnets as
Prototyping Vision-Based Classifiers in Constrained Environments
Prototyping Vision-Based Classifiers in Constrained Environments Ted Hromadka 1 and Cameron Hunt 2 1, 2 SOFWERX (DEFENSEWERX, Inc.) Presented at GTC 2018 Company Overview SM UNCLASSIFIED 2 Capabilities
Supplemental material of Robust Physical-World Attacks on Deep Learning Visual Classification"
Supplemental material of Robust Physical-World Attacks on Deep Learning Visual Classification" Kevin Eykholt 1, Ivan Evtimov *2, Earlence Fernandes 2, Bo Li 3, Amir Rahmati 4, Chaowei Xiao 1, Atul Prakash
Adversarial Examples and Adversarial Training. Ian Goodfellow, OpenAI Research Scientist Presentation at Quora,
Adversarial Examples and Adversarial Training Ian Goodfellow, OpenAI Research Scientist Presentation at Quora, 2016-08-04 In this presentation Intriguing Properties of Neural Networks Szegedy et al, 2013
Automatic point-of-interest image cropping via ensembled convolutionalization
1 Automatic point-of-interest image cropping via ensembled convolutionalization Andrea Asperti and Pietro Battilana University of Bologna Department of informatics: Science and Engineering (DISI) Abstract
Target Recognition and Tracking based on Data Fusion of Radar and Infrared Image Sensors
Target Recognition and Tracking based on Data Fusion of Radar and Infrared Image Sensors Jie YANG Zheng-Gang LU Ying-Kai GUO Institute of Image rocessing & Recognition, Shanghai Jiao-Tong University, China
WITH continuous miniaturization of silicon technology
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. X., X. 8, MONTH 20XX 1 Leveraging expert feature knowledge for predicting image aesthetics Michal Kucer, Student Member, IEEE, Alexander C. Loui, Fellow, IEEE,
Lecture 23 Deep Learning: Segmentation
Lecture 23 Deep Learning: Segmentation COS 429: Computer Vision Thanks: most of these slides shamelessly adapted from Stanford CS231n: Convolutional Neural Networks for Visual Recognition Fei-Fei Li, Andrej
On the Use of Fully Convolutional Networks on Evaluation of Infrared Breast Image Segmentations
17º WIM - Workshop de Informática Médica On the Use of Fully Convolutional Networks on Evaluation of Infrared Breast Image Segmentations Rafael H. C. de Melo, Aura Conci, Cristina Nader Vasconcelos Computer
Defense Against the Dark Arts: Machine Learning Security and Privacy. Ian Goodfellow, Staff Research Scientist, Google Brain BayLearn 2017
Defense Against the Dark Arts: Machine Learning Security and Privacy Ian Goodfellow, Staff Research Scientist, Google Brain BayLearn 2017 An overview of a field This presentation summarizes the work of
Adversarial Examples and Adversarial Training. Ian Goodfellow, OpenAI Research Scientist Presentation at HORSE 2016 London,
Adversarial Examples and Adversarial Training Ian Goodfellow, OpenAI Research Scientist Presentation at HORSE 2016 London, 2016-09-19 In this presentation Intriguing Properties of Neural Networks Szegedy
Spatial Transformer Networks
Spatial Transformer Networks Kaichun Mo Shanghai Jiao Tong University daerduomkch@sjtueducn July 28, 2015 Kaichun Mo (SJTU@Cornell) STN July 28, 2015 1 / 29 Overview 1 Spatial Transformer Network 2 Caffe:
Derek Allman a, Austin Reiter b, and Muyinatu Bell a,c
Exploring the effects of transducer models when training convolutional neural networks to eliminate reflection artifacts in experimental photoacoustic images Derek Allman a, Austin Reiter b, and Muyinatu
A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics
A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics Ossia, SA; Shamsabadi, AS; Taheri, A; Rabiee, HR; Lane, N; Haddadi, H The Author(s) 2017 For additional information about this
Efficient Deep Learning in Communications
Fraunhofer Image Processing Heinrich Hertz Institute Efficient Deep Learning in Communications Dr. Wojciech Samek Fraunhofer HHI, Machine Learning Group Fraunhofer Heinrich Hertz Institute, Einsteinufer
A Deep-Learning-Based Fashion Attributes Detection Model
A Deep-Learning-Based Fashion Attributes Detection Model Menglin Jia Yichen Zhou Mengyun Shi Bharath Hariharan Cornell University {mj493, yz888, ms2979}@cornell.edu, harathh@cs.cornell.edu 1 Introduction
arxiv: v1 [cs.cv] 15 Apr 2016
![arxiv: v1 [cs.cv] 15 Apr 2016](/thumbs/80/81898948.jpg "arxiv: v1 [cs.cv] 15 Apr 2016")
High-performance Semantic Segmentation Using Very Deep Fully Convolutional Networks arxiv:1604.04339v1 [cs.cv] 15 Apr 2016 Zifeng Wu, Chunhua Shen, Anton van den Hengel The University of Adelaide, SA 5005,
Deep Multispectral Semantic Scene Understanding of Forested Environments using Multimodal Fusion
Deep Multispectral Semantic Scene Understanding of Forested Environments using Multimodal Fusion Abhinav Valada, Gabriel L. Oliveira, Thomas Brox, and Wolfram Burgard Department of Computer Science, University
Comparison of Google Image Search and ResNet Image Classification Using Image Similarity Metrics
University of Arkansas, Fayetteville ScholarWorks@UARK Computer Science and Computer Engineering Undergraduate Honors Theses Computer Science and Computer Engineering 5-2018 Comparison of Google Image
A TWO-PART PREDICTIVE CODER FOR MULTITASK SIGNAL COMPRESSION. Scott Deeann Chen and Pierre Moulin
A TWO-PART PREDICTIVE CODER FOR MULTITASK SIGNAL COMPRESSION Scott Deeann Chen and Pierre Moulin University of Illinois at Urbana-Champaign Department of Electrical and Computer Engineering 5 North Mathews
Robust Segmentation of Freight Containers in Train Monitoring Videos
Robust Segmentation of Freight Containers in Train Monitoring Videos Qing-Jie Kong,, Avinash Kumar, Narendra Ahuja, and Yuncai Liu Department of Electrical and Computer Engineering University of Illinois
THE aesthetic quality of an image is judged by commonly
1 Image Aesthetic Assessment: An Experimental Survey Yubin Deng, Chen Change Loy, Member, IEEE, and Xiaoou Tang, Fellow, IEEE arxiv:1610.00838v2 [cs.cv] 20 Apr 2017 Abstract This survey aims at reviewing
ROAD RECOGNITION USING FULLY CONVOLUTIONAL NEURAL NETWORKS
Bulletin of the Transilvania University of Braşov Vol. 10 (59) No. 2-2017 Series I: Engineering Sciences ROAD RECOGNITION USING FULLY CONVOLUTIONAL NEURAL NETWORKS E. HORVÁTH 1 C. POZNA 2 Á. BALLAGI 3
Tokyo Institute of Technology School of Engineering Bachelor Thesis. Real-Time Tennis Ball Speed Analysis System Based on Image Processing
Tokyo Institute of Technology School of Engineering Bachelor Thesis Real-Time Tennis Ball Speed Analysis System Based on Image Processing Supervisor: Associate Prof. Kenji Kise February, 2016 Submitter
arxiv: v1 [cs.cv] 19 Jun 2017
![arxiv: v1 [cs.cv] 19 Jun 2017](/thumbs/80/81286554.jpg "arxiv: v1 [cs.cv] 19 Jun 2017")
Satellite Imagery Feature Detection using Deep Convolutional Neural Network: A Kaggle Competition Vladimir Iglovikov True Accord iglovikov@gmail.com Sergey Mushinskiy Open Data Science cepera.ang@gmail.com
Suneel Marthi Jose Luis Contreras. June 11, 2018 Berlin Buzzwords, Berlin, Germany
Large Scale Landuse Classification of Satellite Imagery Suneel Marthi Jose Luis Contreras June 11, 2018 Berlin Buzzwords, Berlin, Germany 1 Agenda Introduction Satellite Image Data Description Cloud Classification
Ben Baker. Sponsored by:
Ben Baker Sponsored by: Background Agenda GPU Computing Digital Image Processing at FamilySearch Potential GPU based solutions Performance Testing Results Conclusions and Future Work 2 CPU vs. GPU Architecture
THE aesthetic quality of an image is judged by commonly
1 Image Aesthetic Assessment: An Experimental Survey Yubin Deng, Chen Change Loy, Member, IEEE, and Xiaoou Tang, Fellow, IEEE arxiv:1610.00838v1 [cs.cv] 4 Oct 2016 Abstract This survey aims at reviewing
Revolutionize the Service Industries with AI 2016 Service Robot
Revolutionize the Service Industries with AI 2016 Service Robot Clever-m 632 Robot Intelligence Laboratory Jonathan.Xu Standing Vice Director Outline 1 Industry Trends 2 States of Service Robot 3 Powered
Learning to Predict Indoor Illumination from a Single Image. Chih-Hui Ho
Learning to Predict Indoor Illumination from a Single Image Chih-Hui Ho 1 Outline Introduction Method Overview LDR Panorama Light Source Detection Panorama Recentering Warp Learning From LDR Panoramas
Visualizing and Understanding. Fei-Fei Li & Justin Johnson & Serena Yeung. Lecture 12 -
Lecture 12: Visualizing and Understanding Lecture 12-1 May 16, 2017 Administrative Milestones due tonight on Canvas, 11:59pm Midterm grades released on Gradescope this week A3 due next Friday, 5/26 HyperQuest
یادآوری: خالصه CNN. ConvNet
1 ConvNet یادآوری: خالصه CNN شبکه عصبی کانولوشنال یا Convolutional Neural Networks یا نوعی از شبکههای عصبی عمیق مدل یادگیری آن باناظر.اصالح وزنها با الگوریتم back-propagation مناسب برای داده های حجیم و
Learning Rich Features for Image Manipulation Detection
Learning Rich Features for Image Manipulation Detection Peng Zhou Xintong Han Vlad I. Morariu Larry S. Davis University of Maryland, College Park Adobe Research pengzhou@umd.edu {xintong,lsd}@umiacs.umd.edu
Understanding Convolution for Semantic Segmentation
Understanding Convolution for Semantic Segmentation Panqu Wang 1, Pengfei Chen 1, Ye Yuan 2, Ding Liu 3, Zehua Huang 1, Xiaodi Hou 1, Garrison Cottrell 4 1 TuSimple, 2 Carnegie Mellon University, 3 University
DeepUNet: A Deep Fully Convolutional Network for Pixel-level Sea-Land Segmentation
DeepUNet: A Deep Fully Convolutional Network for Pixellevel SeaLand Segmentation Ruirui Li, Wenjie Liu, Lei Yang, Shihao Sun, Wei Hu*, Fan Zhang, Senior Member, IEEE, Wei Li, Senior Member, IEEE Beijing
arxiv: v2 [cs.cv] 25 Apr 2018
![arxiv: v2 [cs.cv] 25 Apr 2018](/thumbs/89/99383977.jpg "arxiv: v2 [cs.cv] 25 Apr 2018")
Driver Gaze Zone Estimation using Convolutional Neural Networks: A General Framework and Ablative Analysis arxiv:1802.02690v2 [cs.cv] 25 Apr 2018 Sourabh Vora, Akshay Rangesh, and Mohan M. Trivedi Abstract
VARIOUS METHODS IN DIGITAL IMAGE PROCESSING. S.Selvaragini 1, E.Venkatesan 2. BIST, BIHER,Bharath University, Chennai-73
Volume 116 No. 16 2017, 265-269 ISSN: 1311-8080 (printed version); ISSN: 1314-3395 (on-line version) url: http://www.ijpam.eu ijpam.eu VARIOUS METHODS IN DIGITAL IMAGE PROCESSING S.Selvaragini 1, E.Venkatesan
Development of Image Processing Tools for Analysis of Laser Deposition Experiments
Development of Image Processing Tools for Analysis of Laser Deposition Experiments Todd Sparks Department of Mechanical and Aerospace Engineering University of Missouri, Rolla Abstract Microscopical metallography
THE problem of automating the solving of
CS231A FINAL PROJECT, JUNE 2016 1 Solving Large Jigsaw Puzzles L. Dery and C. Fufa Abstract This project attempts to reproduce the genetic algorithm in a paper entitled A Genetic Algorithm-Based Solver
Understanding Convolution for Semantic Segmentation
Understanding Convolution for Semantic Segmentation Panqu Wang 1, Pengfei Chen 1, Ye Yuan 2, Ding Liu 3, Zehua Huang 1, Xiaodi Hou 1, Garrison Cottrell 4 1 TuSimple, 2 Carnegie Mellon University, 3 University
Semantic Segmented Style Transfer Kevin Yang* Jihyeon Lee* Julia Wang* Stanford University kyang6
Semantic Segmented Style Transfer Kevin Yang* Jihyeon Lee* Julia Wang* Stanford University kyang6 Stanford University jlee24 Stanford University jwang22 Abstract Inspired by previous style transfer techniques
RECONFIGURABLE SLAM UTILISING FUZZY REASONING
RECONFIGURABLE SLAM UTILISING FUZZY REASONING Dr. Affan Shaukat Abhinav Bajpai Prof Yang Gao 13th Symposium on Advanced Space Technologies in Robotics and Automation ASTRA 2015 11-13 May ESA/ESTEC, Noordwijk,
arxiv: v1 [cs.cv] 25 Sep 2018
![arxiv: v1 [cs.cv] 25 Sep 2018](/thumbs/87/95037815.jpg "arxiv: v1 [cs.cv] 25 Sep 2018")
Satellite Imagery Multiscale Rapid Detection with Windowed Networks Adam Van Etten In-Q-Tel CosmiQ Works avanetten@iqt.org arxiv:1809.09978v1 [cs.cv] 25 Sep 2018 Abstract Detecting small objects over large
Deep Learning for Launching and Mitigating Wireless Jamming Attacks
Deep Learning for Launching and Mitigating Wireless Jamming Attacks Tugba Erpek, Yalin E. Sagduyu, and Yi Shi arxiv:1807.02567v2 [cs.ni] 13 Dec 2018 Abstract An adversarial machine learning approach is
LANDMARK recognition is an important feature for
1 NU-LiteNet: Mobile Landmark Recognition using Convolutional Neural Networks Chakkrit Termritthikun, Surachet Kanprachar, Paisarn Muneesawang arxiv:1810.01074v1 [cs.cv] 2 Oct 2018 Abstract The growth
Automatic Ground Truth Generation of Camera Captured Documents Using Document Image Retrieval
Automatic Ground Truth Generation of Camera Captured Documents Using Document Image Retrieval Sheraz Ahmed, Koichi Kise, Masakazu Iwamura, Marcus Liwicki, and Andreas Dengel German Research Center for
arxiv: v1 [cs.cv] 12 Jul 2017
![arxiv: v1 [cs.cv] 12 Jul 2017](/thumbs/79/79524156.jpg "arxiv: v1 [cs.cv] 12 Jul 2017")
NO Need to Worry about Adversarial Examples in Object Detection in Autonomous Vehicles Jiajun Lu, Hussein Sibai, Evan Fabry, David Forsyth University of Illinois at Urbana Champaign {jlu23, sibai2, efabry2,
Wadehra Kartik, Kathpalia Mukul, Bahl Vasudha, International Journal of Advance Research, Ideas and Innovations in Technology
ISSN: 2454-132X Impact factor: 4.295 (Volume 4, Issue 1) Available online at www.ijariit.com Hand Detection and Gesture Recognition in Real-Time Using Haar-Classification and Convolutional Neural Networks
Video Object Segmentation with Re-identification
Video Object Segmentation with Re-identification Xiaoxiao Li, Yuankai Qi, Zhe Wang, Kai Chen, Ziwei Liu, Jianping Shi Ping Luo, Chen Change Loy, Xiaoou Tang The Chinese University of Hong Kong, SenseTime
Tiny ImageNet Challenge Investigating the Scaling of Inception Layers for Reduced Scale Classification Problems
Tiny ImageNet Challenge Investigating the Scaling of Inception Layers for Reduced Scale Classification Problems Emeric Stéphane Boigné eboigne@stanford.edu Jan Felix Heyse heyse@stanford.edu Abstract Scaling
Deep Neural Network Architectures for Modulation Classification
Deep Neural Network Architectures for Modulation Classification Xiaoyu Liu, Diyu Yang, and Aly El Gamal School of Electrical and Computer Engineering Purdue University Email: {liu1962, yang1467, elgamala}@purdue.edu
Chapter 2 Distributed Consensus Estimation of Wireless Sensor Networks
Chapter 2 Distributed Consensus Estimation of Wireless Sensor Networks Recently, consensus based distributed estimation has attracted considerable attention from various fields to estimate deterministic
Introduction to Machine Learning
Introduction to Machine Learning Deep Learning Barnabás Póczos Credits Many of the pictures, results, and other materials are taken from: Ruslan Salakhutdinov Joshua Bengio Geoffrey Hinton Yann LeCun 2
Communications Planner for Operational and Simulation Effects With Realism (COMPOSER)
Communications Planner for Operational and Simulation Effects With Realism (COMPOSER) Alan J. Scrime CERDEC Chief, Spectrum Analysis & Frequency Management Branch (732) 427-6346, alan.scrime@us.army.mil
RESEARCH TRENDS IN HYPERSPECTRAL IMAGERY DATA
International Journal of Computer Engineering & Technology (IJCET) Volume 10, Issue 2, March-April 2019, pp. 67-73, Article ID: IJCET_10_02_008 Available online at http://www.iaeme.com/ijcet/issues.asp?jtype=ijcet&vtype=10&itype=2
Artistic Image Colorization with Visual Generative Networks
Artistic Image Colorization with Visual Generative Networks Final report Yuting Sun ytsun@stanford.edu Yue Zhang zoezhang@stanford.edu Qingyang Liu qnliu@stanford.edu 1 Motivation Visual generative models,
GLOBAL BLUR ASSESSMENT AND BLURRED REGION DETECTION IN NATURAL IMAGES
GLOBAL BLUR ASSESSMENT AND BLURRED REGION DETECTION IN NATURAL IMAGES Loreta A. ŞUTA, Mircea F. VAIDA Technical University of Cluj-Napoca, 26-28 Baritiu str. Cluj-Napoca, Romania Phone: +40-264-401226,
Consistent Comic Colorization with Pixel-wise Background Classification
Consistent Comic Colorization with Pixel-wise Background Classification Sungmin Kang KAIST Jaegul Choo Korea University Jaehyuk Chang NAVER WEBTOON Corp. Abstract Comic colorization is a time-consuming
Color Constancy Using Standard Deviation of Color Channels
2010 International Conference on Pattern Recognition Color Constancy Using Standard Deviation of Color Channels Anustup Choudhury and Gérard Medioni Department of Computer Science University of Southern
Binary Neural Network and Its Implementation with 16 Mb RRAM Macro Chip
Binary Neural Network and Its Implementation with 16 Mb RRAM Macro Chip Assistant Professor of Electrical Engineering and Computer Engineering shimengy@asu.edu http://faculty.engineering.asu.edu/shimengyu/
arxiv: v3 [cs.cv] 18 Dec 2018
![arxiv: v3 [cs.cv] 18 Dec 2018](/thumbs/96/128789272.jpg "arxiv: v3 [cs.cv] 18 Dec 2018")
Video Colorization using CNNs and Keyframes extraction: An application in saving bandwidth Ankur Singh 1 Anurag Chanani 2 Harish Karnick 3 arxiv:1812.03858v3 [cs.cv] 18 Dec 2018 Abstract In this paper,
To Infinity And Beyond. Computer Vision for Astronomy
To Infinity And Beyond Computer Vision for Astronomy Ryan Fox ryan@foxrow.com @ryan_fox foxrow.com 1. Image Processing 2. Computer Vision 3. To Infinity and Beyond How computers see How computers see 006
To Post or Not To Post: Using CNNs to Classify Social Media Worthy Images
To Post or Not To Post: Using CNNs to Classify Social Media Worthy Images Lauren Blake Stanford University lblake@stanford.edu Abstract This project considers the feasibility for CNN models to classify
신경망기반자동번역기술. Konkuk University Computational Intelligence Lab. 김강일
신경망기반자동번역기술 Konkuk University Computational Intelligence Lab. http://ci.konkuk.ac.kr kikim01@kunkuk.ac.kr 김강일 Index Issues in AI and Deep Learning Overview of Machine Translation Advanced Techniques in
Novel Methods for Microscopic Image Processing, Analysis, Classification and Compression
Novel Methods for Microscopic Image Processing, Analysis, Classification and Compression Ph.D. Defense by Alexander Suhre Supervisor: Prof. A. Enis Çetin March 11, 2013 Outline Storage Analysis Image Acquisition
Automatic Playlist Generation
Automatic Generation Xingting Gong and Xu Chen Stanford University gongx@stanford.edu xchen91@stanford.edu I. Introduction Digital music applications have become an increasingly popular means of listening
CT parameter studies for porous metal samples. Sören R. Lindemann Daimler AG Werk Untertürkheim
CT parameter studies for porous metal samples Sören R. Lindemann Daimler AG Werk Untertürkheim Where do we stand and what are we looking for? small material samples (high absorption coefficient, low porosity)
Advanced Dimensional Management LLC
Index for Tolerance Stackup, Tolerance Analysis and Tolerancing Manual: Plus and Minus Edition Accuracy and precision 3, 19 Ambiguity 20 ASME Y14.5M-1994 2 rule #1 2 Assembly Process 20, 28, 79-81 Assembly
The Rise of AI: Status, Thresholds, Attack Surfaces
The Rise of AI: Status, Thresholds, Attack Surfaces Subbarao Kambhampati The Many Intelligences.. H u m a n s Perceptual & Manipulation intelligence that seem to come naturally to us Image recognition;
Scene Perception based on Boosting over Multimodal Channel Features
Scene Perception based on Boosting over Multimodal Channel Features Arthur Costea Image Processing and Pattern Recognition Research Center Technical University of Cluj-Napoca Research Group Technical University
Exercise 2: Hodgkin and Huxley model
Exercise 2: Hodgkin and Huxley model Expected time: 4.5h To complete this exercise you will need access to MATLAB version 6 or higher (V5.3 also seems to work), and the Hodgkin-Huxley simulator code. At
1.Discuss the frequency domain techniques of image enhancement in detail.
1.Discuss the frequency domain techniques of image enhancement in detail. Enhancement In Frequency Domain: The frequency domain methods of image enhancement are based on convolution theorem. This is represented
Challenges for Deep Scene Understanding
Challenges for Deep Scene Understanding BoleiZhou MIT Bolei Zhou Hang Zhao Xavier Puig Sanja Fidler (UToronto) Adela Barriuso Aditya Khosla Antonio Torralba Aude Oliva Objects in the Scene Context Challenge
LifeCLEF Bird Identification Task 2016
LifeCLEF Bird Identification Task 2016 The arrival of deep learning Alexis Joly, Inria Zenith Team, Montpellier, France Hervé Glotin, Univ. Toulon, UMR LSIS, Institut Universitaire de France Hervé Goëau,
A New Fake Iris Detection Method
A New Fake Iris Detection Method Xiaofu He 1, Yue Lu 1, and Pengfei Shi 2 1 Department of Computer Science and Technology, East China Normal University, Shanghai 200241, China {xfhe,ylu}@cs.ecnu.edu.cn