Reconstructive Sparse Code Transfer for Contour Detection and Semantic Labeling
|
|
- Bryan Dean
- 6 years ago
- Views:
Transcription
1 Reconstructive Sparse Code Transfer for Contour Detection and Semantic Labeling Michael Maire 1,2 Stella X. Yu 3 Pietro Perona 2 1 TTI Chicago 2 California Institute of Technology 3 University of California at Berkeley / ICSI
2 Motivation: Deep Representations
3 Image Classification Motivation: Deep Representations [Krizhevsky, Sutskever, and Hinton, NIPS 2012]
4 Image Classification Motivation: Deep Representations [Krizhevsky, Sutskever, and Hinton, NIPS 2012] [Bo, Ren, and Fox, CVPR 2013]
5 Image Classification Motivation: Deep Representations [Krizhevsky, Sutskever, and Hinton, NIPS 2012] [Bo, Ren, and Fox, CVPR 2013] Object Detection [Girshick, Donahue, Darrell, and Malik, CVPR 2014]
6 Deep Representations for Semantic Labeling classify image detect objects
7 Deep Representations for Semantic Labeling classify image label every pixel detect objects
8 Contour Detection: Special Case { 0 if in region interior 1 if on region boundary
9 Contour Detection: Special Case
10 Contour Detection: Special Case contour detection serves as foundation for: segmentation, object proposals
11 Semantic Labeling Strategy predict patch labels from a spatially localized multilayer slice of a deep representation
12 Semantic Labeling Strategy predict patch labels from a spatially localized multilayer slice of a deep representation generalization of sparse reconstruction
13 Multipath Sparse Coding [Bo, Ren, and Fox, CVPR 2013]
14 [Bo, Ren, and Fox, CVPR 2013]
15 [Bo, Ren, and Fox, CVPR 2013]
16 [Bo, Ren, and Fox, CVPR 2013]
17
18 patch descriptor (sparse)
19 use patch descriptors in a sparse reconstructive setting patch descriptor (sparse)
20 Sparse Coding & Reconstruction Image
21 Sparse Coding & Reconstruction Image Patch Dictionary Batch OMP
22 Sparse Coding & Reconstruction Image Patch Dictionary Batch OMP = z i,0 + z i, z i,l z i 0 K x i d 0 d 1 d L
23 Sparse Coding & Reconstruction Image Patch Dictionary Sparse Codes Batch OMP coefficients pixels = z i,0 + z i, z i,l z i 0 K x i d 0 d 1 d L
24 Sparse Coding & Reconstruction Image Patch Dictionary Sparse Codes Batch OMP coefficients pixels = z i,0 + z i, z i,l z i 0 K x i d 0 d 1 d L
25 Sparse Coding & Reconstruction Image Patch Dictionary Sparse Codes Batch OMP coefficients pixels
26 Sparse Coding & Reconstruction Image Patch Dictionary Sparse Codes Batch OMP coefficients pixels Sparse Codes coefficients pixels
27 Sparse Coding & Reconstruction Image Patch Dictionary Sparse Codes Batch OMP coefficients pixels coefficients Sparse Codes pixels Patch Dictionary Reconstruction
28 Sparse Coding & Reconstruction Image Patch Dictionary Sparse Codes Batch OMP coefficients pixels coefficients Sparse Codes pixels Patch Dictionary Reconstruction Image
29 Reconstructive Sparse Code Transfer Sparse Codes coefficients pixels
30 Reconstructive Sparse Code Transfer Rectified Sparse Codes coefficients pixels [ ] z i max(zi T, 0), max( zi T T, 0), 1
31 Reconstructive Sparse Code Transfer Rectified Sparse Codes Transfer Dictionary coefficients pixels [ ] z i max(zi T, 0), max( zi T T, 0), 1 Reconstruction
32 Reconstructive Sparse Code Transfer Rectified Sparse Codes Transfer Dictionary coefficients pixels [ ] z i max(zi T, 0), max( zi T T, 0), 1 Reconstruction Contour Detection
33 Sparse Coding: Generative Training
34 Sparse Coding: Generative Training sample patches X = [x 0, x 1,...] from training images
35 Sparse Coding: Generative Training sample patches X = [x 0, x 1,...] from training images learn: sparse representations Z = [z0, z 1,...] dictionary D = [d0, d 1,..., d L 1 ]
36 Sparse Coding: Generative Training sample patches X = [x 0, x 1,...] from training images learn: sparse representations Z = [z0, z 1,...] dictionary D = [d0, d 1,..., d L 1 ] MI-KSVD finds an approximate solution to: [ L 1 L 1 X DZ 2 F + λ argmin D, Z i=0 j=0,j i s.t. i, d i 2 = 1 and n, z n 0 K d T i d j ]
37 Sparse Coding: Generative Training sample patches X = [x 0, x 1,...] from training images learn: sparse representations Z = [z0, z 1,...] dictionary D = [d0, d 1,..., d L 1 ] MI-KSVD finds an approximate solution to: [ L 1 L 1 X DZ 2 F + λ argmin D, Z i=0 j=0,j i s.t. i, d i 2 = 1 and n, z n 0 K encode patch x R m m c as z R L : d T i d j ] argmin x Dz 2 z s.t. z 0 K
38 Transfer Dictionary: Discriminative Training
39 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c
40 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( )
41 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h
42 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h patch-level discriminative training:
43 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h patch-level discriminative training: sample codes {z0, z 1,...} and corresponding groundtruth patches {g 0, g 1,...}
44 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h patch-level discriminative training: sample codes {z0, z 1,...} and corresponding groundtruth patches {g 0, g 1,...} split F ( ) into independently trained predictors [f 0, f 1,..., f (m 2 h 1)], one for each output element
45 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h patch-level discriminative training: sample codes {z0, z 1,...} and corresponding groundtruth patches {g 0, g 1,...} split F ( ) into independently trained predictors [f 0, f 1,..., f (m 2 h 1)], one for each output element we want: f j (z i ) g i [j] i, j
46 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h patch-level discriminative training: sample codes {z0, z 1,...} and corresponding groundtruth patches {g 0, g 1,...} split F ( ) into independently trained predictors [f 0, f 1,..., f (m 2 h 1)], one for each output element we want: f j (z i ) g i [j] i, j train each fj ( ) using logistic regression
47 Transfer Dictionary: Discriminative Training D maps sparse vector z R L to patch P R m m c replace D with function F ( ) F (z) predicts groundtruth labeling G R m m h patch-level discriminative training: sample codes {z0, z 1,...} and corresponding groundtruth patches {g 0, g 1,...} split F ( ) into independently trained predictors [f 0, f 1,..., f (m 2 h 1)], one for each output element we want: f j (z i ) g i [j] i, j train each fj ( ) using logistic regression replace z with concatenated multipath codes
48
49 Multiple Scales
50 Multiple Scales Layer 1 Dictionaries 5x5, 64 atoms 11x11, 64 atoms 5x5, 512 atoms 11x11, 512 atoms 21x21, 512 atoms 31x31, 512 atoms
51 Multiple Scales Layer 1 Dictionaries 5x5, 64 atoms 11x11, 64 atoms 5x5, 512 atoms 11x11, 512 atoms 21x21, 512 atoms 31x31, 512 atoms pooling Layer 2 Dictionaries 5x5, 512 atoms 5x5, 512 atoms
52 Multiple Scales Layer 1 Dictionaries 5x5, 64 atoms 11x11, 64 atoms 5x5, 512 atoms 11x11, 512 atoms 21x21, 512 atoms 31x31, 512 atoms pooling = Layer 2 Dictionaries 5x5, 512 atoms 5x5, 512 atoms rectify, upsample, concatenate sparse activation maps
53 Multiple Scales Layer 1 Dictionaries 5x5, 64 atoms 11x11, 64 atoms 5x5, 512 atoms 11x11, 512 atoms 21x21, 512 atoms 31x31, 512 atoms pooling = Layer 2 Dictionaries 5x5, 512 atoms 5x5, 512 atoms rectify, upsample, concatenate sparse activation maps dimensions Sparse Representation
54 Multiple Scales Layer 1 Dictionaries 5x5, 64 atoms 11x11, 64 atoms 5x5, 512 atoms 11x11, 512 atoms 21x21, 512 atoms 31x31, 512 atoms pooling = Layer 2 Dictionaries 5x5, 512 atoms 5x5, 512 atoms rectify, upsample, concatenate sparse activation maps = dimensions Contour Reconstruction Sparse Representation
55 Contour Detection Groundtruth [Martin, Fowlkes, Tal, and Malik, ICCV 2001]
56 Contour Detection Results
57 Contour Detection Results
58 Contour Detection Performance
59 Contour Detection Performance
60 Contour Detection Performance Performance Metric Hand-Designed Spectral ODS F OIS F AP Features? Filters? Globalization? Human Structured Edges yes no no local SCG (color) no yes no Sparse Code Transfer Layers no no no Sparse Code Transfer Layer no no no local SCG (gray) no yes no multiscale Pb yes yes no Canny Edge Detector yes yes no global SCG (color) yes yes yes global Pb + UCM yes yes yes + UCM global Pb yes yes yes Sparse Code Transfer: Performance competitive with top approaches Both representation and classifier are learned Free from reliance on hand-designed features or filters
61 Texture and Network Depth Layer 1 Layers 1+2
62 Texture and Network Depth Layer 1 Layers 1+2
63 Semantic Labeling hair skin background Labeled Faces in the Wild Dataset [Kae, Sohn, Lee, Learned-Miller, CVPR 2013]
64 Summary
65 Summary Semantic labeling using spatially localized slices of deep representations
66 Summary Semantic labeling using spatially localized slices of deep representations Sparse reconstruction generalized to multipath networks
67 Summary Semantic labeling using spatially localized slices of deep representations Sparse reconstruction generalized to multipath networks Transfer learning perspective: Generatively trained patch representation Task-specific discriminately trained transfer classifier
68 Summary Semantic labeling using spatially localized slices of deep representations Sparse reconstruction generalized to multipath networks Transfer learning perspective: Generatively trained patch representation Task-specific discriminately trained transfer classifier Pure learning for high performance contour detection
69 Summary Semantic labeling using spatially localized slices of deep representations Sparse reconstruction generalized to multipath networks Transfer learning perspective: Generatively trained patch representation Task-specific discriminately trained transfer classifier Pure learning for high performance contour detection Contours obtained as a byproduct of deep representations
70 Summary Semantic labeling using spatially localized slices of deep representations Sparse reconstruction generalized to multipath networks Transfer learning perspective: Generatively trained patch representation Task-specific discriminately trained transfer classifier Pure learning for high performance contour detection Contours obtained as a byproduct of deep representations Thank You!
arxiv: v1 [cs.cv] 16 Oct 2014
Reconstructive Sparse Code Transfer for Contour Detection and Semantic Labeling Michael Maire 1,2, Stella X. Yu 3, and Pietro Perona 2 arxiv:1410.4521v1 [cs.cv] 16 Oct 2014 1 TTI Chicago 2 California Institute
More informationEdge Detection Using Convolutional Neural Network
Edge Detection Using Convolutional Neural Network Ruohui Wang (B) Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong, China wr013@ie.cuhk.edu.hk Abstract. In this work,
More informationDeepEdge: A Multi-Scale Bifurcated Deep Network for Top-Down Contour Detection
DeepEdge: A Multi-Scale Bifurcated Deep Network for Top-Down Contour Detection Gedas Bertasius University of Pennsylvania gberta@seas.upenn.edu Jianbo Shi University of Pennsylvania jshi@seas.upenn.edu
More informationLearning the Ecological Statistics of Perceptual Organization
Learning the Ecological Statistics of Perceptual Organization Charless Fowlkes work with David Martin, Xiaofeng Ren and Jitendra Malik at University of California at Berkeley 1 How do ideas from perceptual
More informationLecture 5: Object Detection
Object Detection CSED703R: Deep Learning for Visual Recognition (2017F) Lecture 5: Object Detection Bohyung Han Computer Vision Lab. bhhan@postech.ac.kr 2 Traditional Object Detection Algorithms Region-based
More informationREGION AVERAGE POOLING FOR CONTEXT-AWARE OBJECT DETECTION
REGION AVERAGE POOLING FOR CONTEXT-AWARE OBJECT DETECTION Kingsley Kuan 1, Gaurav Manek 1, Jie Lin 1, Yuan Fang 1, Vijay Chandrasekhar 1,2 Institute for Infocomm Research, A*STAR, Singapore 1 Nanyang Technological
More informationIMAGE SUPER-RESOLUTION BASED ON DICTIONARY LEARNING AND ANCHORED NEIGHBORHOOD REGRESSION WITH MUTUAL INCOHERENCE
IMAGE SUPER-RESOLUTION BASED ON DICTIONARY LEARNING AND ANCHORED NEIGHBORHOOD REGRESSION WITH MUTUAL INCOHERENCE Yulun Zhang 1, Kaiyu Gu 2, Yongbing Zhang 1, Jian Zhang 3, and Qionghai Dai 1,4 1 Shenzhen
More informationA FRAMEWORK OF EXTRACTING MULTI-SCALE FEATURES USING MULTIPLE CONVOLUTIONAL NEURAL NETWORKS. Kuan-Chuan Peng and Tsuhan Chen
A FRAMEWORK OF EXTRACTING MULTI-SCALE FEATURES USING MULTIPLE CONVOLUTIONAL NEURAL NETWORKS Kuan-Chuan Peng and Tsuhan Chen School of Electrical and Computer Engineering, Cornell University, Ithaca, NY
More informationFast Edge Detection Using Structured Forests
Fast Edge Detection Using Structured Forests Piotr Dollár, C. Lawrence Zitnick [1] Zhihao Li (zhihaol@andrew.cmu.edu) Computer Science Department Carnegie Mellon University Table of contents 1. Introduction
More informationComputer Vision Lecture 16
Computer Vision Lecture 16 Deep Learning Applications 11.01.2017 Bastian Leibe RWTH Aachen http://www.vision.rwth-aachen.de leibe@vision.rwth-aachen.de Announcements Seminar registration period starts
More informationComputer Vision Lecture 16
Announcements Computer Vision Lecture 16 Deep Learning Applications 11.01.2017 Seminar registration period starts on Friday We will offer a lab course in the summer semester Deep Robot Learning Topic:
More informationPart Localization by Exploiting Deep Convolutional Networks
Part Localization by Exploiting Deep Convolutional Networks Marcel Simon, Erik Rodner, and Joachim Denzler Computer Vision Group, Friedrich Schiller University of Jena, Germany www.inf-cv.uni-jena.de Abstract.
More informationDeep learning for object detection. Slides from Svetlana Lazebnik and many others
Deep learning for object detection Slides from Svetlana Lazebnik and many others Recent developments in object detection 80% PASCAL VOC mean0average0precision0(map) 70% 60% 50% 40% 30% 20% 10% Before deep
More informationMultipath Sparse Coding Using Hierarchical Matching Pursuit
Multipath Sparse Coding Using Hierarchical Matching Pursuit Liefeng Bo ISTC-PC Intel Labs liefeng.bo@intel.com Xiaofeng Ren ISTC-PC Intel Labs xren@cs.washington.edu Dieter Fox University of Washington
More informationMultipath Sparse Coding Using Hierarchical Matching Pursuit
2013 IEEE Conference on Computer Vision and Pattern Recognition Multipath Sparse Coding Using Hierarchical Matching Pursuit Liefeng Bo ISTC-PC Intel Labs liefeng.bo@intel.com Xiaofeng Ren ISTC-PC Intel
More informationBoundaries and Sketches
Boundaries and Sketches Szeliski 4.2 Computer Vision James Hays Many slides from Michael Maire, Jitendra Malek Today s lecture Segmentation vs Boundary Detection Why boundaries / Grouping? Recap: Canny
More informationPart III: Affinity Functions for Image Segmentation
Part III: Affinity Functions for Image Segmentation Charless Fowlkes joint work with David Martin and Jitendra Malik at University of California at Berkeley 1 Q: What measurements should we use for constructing
More informationMulti-View 3D Object Detection Network for Autonomous Driving
Multi-View 3D Object Detection Network for Autonomous Driving Xiaozhi Chen, Huimin Ma, Ji Wan, Bo Li, Tian Xia CVPR 2017 (Spotlight) Presented By: Jason Ku Overview Motivation Dataset Network Architecture
More informationDeepContour: A Deep Convolutional Feature Learned by Positive-sharing Loss for Contour Detection
DeepContour: A Deep Convolutional Feature Learned by Positive-sharing Loss for Contour Detection Wei Shen 1, Xinggang Wang 2, Yan Wang 3, Xiang Bai 2, Zhijiang Zhang 1 1 Key Lab of Specialty Fiber Optics
More informationStacked Multiscale Feature Learning for Domain Independent Medical Image Segmentation
Stacked Multiscale Feature Learning for Domain Independent Medical Image Segmentation Ryan Kiros, Karteek Popuri, Dana Cobzas, and Martin Jagersand University of Alberta, Canada Abstract. In this work
More information3D Object Recognition and Scene Understanding from RGB-D Videos. Yu Xiang Postdoctoral Researcher University of Washington
3D Object Recognition and Scene Understanding from RGB-D Videos Yu Xiang Postdoctoral Researcher University of Washington 1 2 Act in the 3D World Sensing & Understanding Acting Intelligent System 3D World
More informationObject detection with CNNs
Object detection with CNNs 80% PASCAL VOC mean0average0precision0(map) 70% 60% 50% 40% 30% 20% 10% Before CNNs After CNNs 0% 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 year Region proposals
More informationEdge Detection. Computer Vision Shiv Ram Dubey, IIIT Sri City
Edge Detection Computer Vision Shiv Ram Dubey, IIIT Sri City Previous two classes: Image Filtering Spatial domain Smoothing, sharpening, measuring texture * = FFT FFT Inverse FFT = Frequency domain Denoising,
More informationComputer Vision Lecture 16
Computer Vision Lecture 16 Deep Learning for Object Categorization 14.01.2016 Bastian Leibe RWTH Aachen http://www.vision.rwth-aachen.de leibe@vision.rwth-aachen.de Announcements Seminar registration period
More informationDiscrete-Continuous Gradient Orientation Estimation for Faster Image Segmentation
Discrete-Continuous Gradient Orientation Estimation for Faster Image Segmentation Michael Donoser and Dieter Schmalstieg Institute for Computer Graphics and Vision Graz University of Technology {donoser,schmalstieg}@icg.tugraz.at
More informationExtended Dictionary Learning : Convolutional and Multiple Feature Spaces
Extended Dictionary Learning : Convolutional and Multiple Feature Spaces Konstantina Fotiadou, Greg Tsagkatakis & Panagiotis Tsakalides kfot@ics.forth.gr, greg@ics.forth.gr, tsakalid@ics.forth.gr ICS-
More informationGlobal Probability of Boundary
Global Probability of Boundary Learning to Detect Natural Image Boundaries Using Local Brightness, Color, and Texture Cues Martin, Fowlkes, Malik Using Contours to Detect and Localize Junctions in Natural
More informationMultipath Sparse Coding Using Hierarchical Matching Pursuit
Multipath Sparse Coding Using Hierarchical Matching Pursuit Liefeng Bo, Xiaofeng Ren ISTC Pervasive Computing, Intel Labs Seattle WA 98195, USA {liefeng.bo,xiaofeng.ren}@intel.com Dieter Fox University
More informationCS 2770: Computer Vision. Edges and Segments. Prof. Adriana Kovashka University of Pittsburgh February 21, 2017
CS 2770: Computer Vision Edges and Segments Prof. Adriana Kovashka University of Pittsburgh February 21, 2017 Edges vs Segments Figure adapted from J. Hays Edges vs Segments Edges More low-level Don t
More informationPerceiving the 3D World from Images and Videos. Yu Xiang Postdoctoral Researcher University of Washington
Perceiving the 3D World from Images and Videos Yu Xiang Postdoctoral Researcher University of Washington 1 2 Act in the 3D World Sensing & Understanding Acting Intelligent System 3D World 3 Understand
More informationContent-Based Image Recovery
Content-Based Image Recovery Hong-Yu Zhou and Jianxin Wu National Key Laboratory for Novel Software Technology Nanjing University, China zhouhy@lamda.nju.edu.cn wujx2001@nju.edu.cn Abstract. We propose
More informationPASCAL Boundaries: A Semantic Boundary Dataset with A Deep Semantic Boundary Detector
PASCAL Boundaries: A Semantic Boundary Dataset with A Deep Semantic Boundary Detector Vittal Premachandran 1, Boyan Bonev 2, Xiaochen Lian 2, and Alan Yuille 1 1 Department of Computer Science, Johns Hopkins
More informationLEARNING BOUNDARIES WITH COLOR AND DEPTH. Zhaoyin Jia, Andrew Gallagher, Tsuhan Chen
LEARNING BOUNDARIES WITH COLOR AND DEPTH Zhaoyin Jia, Andrew Gallagher, Tsuhan Chen School of Electrical and Computer Engineering, Cornell University ABSTRACT To enable high-level understanding of a scene,
More informationDense Volume-to-Volume Vascular Boundary Detection
Dense Volume-to-Volume Vascular Boundary Detection Jameson Merkow 1, David Kriegman 1, Alison Marsden 2, and Zhuowen Tu 1 1 University of California, San Diego. 2 Stanford University Abstract. In this
More informationCS6501: Deep Learning for Visual Recognition. Object Detection I: RCNN, Fast-RCNN, Faster-RCNN
CS6501: Deep Learning for Visual Recognition Object Detection I: RCNN, Fast-RCNN, Faster-RCNN Today s Class Object Detection The RCNN Object Detector (2014) The Fast RCNN Object Detector (2015) The Faster
More informationAffinity CNN: Learning Pixel-Centric Pairwise Relations for Figure/Ground Embedding
Affinity CNN: Learning Pixel-Centric Pairwise Relations for Figure/Ground Embedding Michael Maire TTI Chicago mmaire@ttic.edu Takuya Narihira Sony Corp. takuya.narihira@jp.sony.com Stella X. Yu UC Berkeley
More informationLearning-based Methods in Vision
Learning-based Methods in Vision 16-824 Sparsity and Deep Learning Motivation Multitude of hand-designed features currently in use in vision - SIFT, HoG, LBP, MSER, etc. Even the best approaches, just
More informationarxiv: v1 [cs.cv] 17 May 2016
SemiContour: A Semi-supervised Learning Approach for Contour Detection Zizhao Zhang, Fuyong Xing, Xiaoshuang Shi, Lin Yang University of Florida, Gainesville, FL 32611, USA zizhao@cise.ufl.edu, {f.xing,xsshi2015}@ufl.edu,
More informationLecture 7: Semantic Segmentation
Semantic Segmentation CSED703R: Deep Learning for Visual Recognition (207F) Segmenting images based on its semantic notion Lecture 7: Semantic Segmentation Bohyung Han Computer Vision Lab. bhhanpostech.ac.kr
More informationDeep Neural Networks:
Deep Neural Networks: Part II Convolutional Neural Network (CNN) Yuan-Kai Wang, 2016 Web site of this course: http://pattern-recognition.weebly.com source: CNN for ImageClassification, by S. Lazebnik,
More informationTRANSPARENT OBJECT DETECTION USING REGIONS WITH CONVOLUTIONAL NEURAL NETWORK
TRANSPARENT OBJECT DETECTION USING REGIONS WITH CONVOLUTIONAL NEURAL NETWORK 1 Po-Jen Lai ( 賴柏任 ), 2 Chiou-Shann Fuh ( 傅楸善 ) 1 Dept. of Electrical Engineering, National Taiwan University, Taiwan 2 Dept.
More informationTowards Large-Scale Semantic Representations for Actionable Exploitation. Prof. Trevor Darrell UC Berkeley
Towards Large-Scale Semantic Representations for Actionable Exploitation Prof. Trevor Darrell UC Berkeley traditional surveillance sensor emerging crowd sensor Desired capabilities: spatio-temporal reconstruction
More informationDirect Intrinsics: Learning Albedo-Shading Decomposition by Convolutional Regression
Direct Intrinsics: Learning - Decomposition by Convolutional Regression Takuya Narihira UC Berkeley / ICSI / Sony Corp. takuya.narihira@jp.sony.com Michael Maire TTI Chicago mmaire@ttic.edu Stella X. Yu
More informationProceedings of the International MultiConference of Engineers and Computer Scientists 2018 Vol I IMECS 2018, March 14-16, 2018, Hong Kong
, March 14-16, 2018, Hong Kong , March 14-16, 2018, Hong Kong , March 14-16, 2018, Hong Kong , March 14-16, 2018, Hong Kong TABLE I CLASSIFICATION ACCURACY OF DIFFERENT PRE-TRAINED MODELS ON THE TEST DATA
More informationSingle Image Depth Estimation via Deep Learning
Single Image Depth Estimation via Deep Learning Wei Song Stanford University Stanford, CA Abstract The goal of the project is to apply direct supervised deep learning to the problem of monocular depth
More informationMultiscale Combinatorial Grouping
Multiscale Combinatorial Grouping Pablo Arbela ez1, Jordi Pont-Tuset2, 1 Jonathan T. Barron1 University of California, Berkeley Berkeley, CA 94720 2 Ferran Marques2 Jitendra Malik1 Universitat Polite cnica
More informationClassification of objects from Video Data (Group 30)
Classification of objects from Video Data (Group 30) Sheallika Singh 12665 Vibhuti Mahajan 12792 Aahitagni Mukherjee 12001 M Arvind 12385 1 Motivation Video surveillance has been employed for a long time
More informationSupplementary material for Analyzing Filters Toward Efficient ConvNet
Supplementary material for Analyzing Filters Toward Efficient Net Takumi Kobayashi National Institute of Advanced Industrial Science and Technology, Japan takumi.kobayashi@aist.go.jp A. Orthonormal Steerable
More informationCost-alleviative Learning for Deep Convolutional Neural Network-based Facial Part Labeling
[DOI: 10.2197/ipsjtcva.7.99] Express Paper Cost-alleviative Learning for Deep Convolutional Neural Network-based Facial Part Labeling Takayoshi Yamashita 1,a) Takaya Nakamura 1 Hiroshi Fukui 1,b) Yuji
More informationRich feature hierarchies for accurate object detection and semantic segmentation
Rich feature hierarchies for accurate object detection and semantic segmentation BY; ROSS GIRSHICK, JEFF DONAHUE, TREVOR DARRELL AND JITENDRA MALIK PRESENTER; MUHAMMAD OSAMA Object detection vs. classification
More informationLEARNING A SPARSE DICTIONARY OF VIDEO STRUCTURE FOR ACTIVITY MODELING. Nandita M. Nayak, Amit K. Roy-Chowdhury. University of California, Riverside
LEARNING A SPARSE DICTIONARY OF VIDEO STRUCTURE FOR ACTIVITY MODELING Nandita M. Nayak, Amit K. Roy-Chowdhury University of California, Riverside ABSTRACT We present an approach which incorporates spatiotemporal
More informationREJECTION-BASED CLASSIFICATION FOR ACTION RECOGNITION USING A SPATIO-TEMPORAL DICTIONARY. Stefen Chan Wai Tim, Michele Rombaut, Denis Pellerin
REJECTION-BASED CLASSIFICATION FOR ACTION RECOGNITION USING A SPATIO-TEMPORAL DICTIONARY Stefen Chan Wai Tim, Michele Rombaut, Denis Pellerin Univ. Grenoble Alpes, GIPSA-Lab, F-38000 Grenoble, France ABSTRACT
More informationMachine Learning. MGS Lecture 3: Deep Learning
Dr Michel F. Valstar http://cs.nott.ac.uk/~mfv/ Machine Learning MGS Lecture 3: Deep Learning Dr Michel F. Valstar http://cs.nott.ac.uk/~mfv/ WHAT IS DEEP LEARNING? Shallow network: Only one hidden layer
More informationFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun Presented by Tushar Bansal Objective 1. Get bounding box for all objects
More informationEdge and Texture. CS 554 Computer Vision Pinar Duygulu Bilkent University
Edge and Texture CS 554 Computer Vision Pinar Duygulu Bilkent University Filters for features Previously, thinking of filtering as a way to remove or reduce noise Now, consider how filters will allow us
More informationDeep Crisp Boundaries
Deep Crisp Boundaries Yupei Wang 1,2, Xin Zhao 1,2, Kaiqi Huang 1,2,3 1 CRIPAC & NLPR, CASIA 2 University of Chinese Academy of Sciences 3 CAS Center for Excellence in Brain Science and Intelligence Technology
More informationarxiv: v1 [cs.cv] 20 Dec 2016
End-to-End Pedestrian Collision Warning System based on a Convolutional Neural Network with Semantic Segmentation arxiv:1612.06558v1 [cs.cv] 20 Dec 2016 Heechul Jung heechul@dgist.ac.kr Min-Kook Choi mkchoi@dgist.ac.kr
More informationMicroscopy Cell Counting with Fully Convolutional Regression Networks
Microscopy Cell Counting with Fully Convolutional Regression Networks Weidi Xie, J. Alison Noble, Andrew Zisserman Department of Engineering Science, University of Oxford,UK Abstract. This paper concerns
More informationComputer Vision I. Announcement. Corners. Edges. Numerical Derivatives f(x) Edge and Corner Detection. CSE252A Lecture 11
Announcement Edge and Corner Detection Slides are posted HW due Friday CSE5A Lecture 11 Edges Corners Edge is Where Change Occurs: 1-D Change is measured by derivative in 1D Numerical Derivatives f(x)
More informationSupervised texture detection in images
Supervised texture detection in images Branislav Mičušík and Allan Hanbury Pattern Recognition and Image Processing Group, Institute of Computer Aided Automation, Vienna University of Technology Favoritenstraße
More informationRecap from Monday. Visualizing Networks Caffe overview Slides are now online
Recap from Monday Visualizing Networks Caffe overview Slides are now online Today Edges and Regions, GPB Fast Edge Detection Using Structured Forests Zhihao Li Holistically-Nested Edge Detection Yuxin
More informationObject Detection Based on Deep Learning
Object Detection Based on Deep Learning Yurii Pashchenko AI Ukraine 2016, Kharkiv, 2016 Image classification (mostly what you ve seen) http://tutorial.caffe.berkeleyvision.org/caffe-cvpr15-detection.pdf
More informationVisual Material Traits Recognizing Per-Pixel Material Context
Visual Material Traits Recognizing Per-Pixel Material Context Gabriel Schwartz gbs25@drexel.edu Ko Nishino kon@drexel.edu 1 / 22 2 / 22 Materials as Visual Context What tells us this road is unsafe? 3
More informationIndustrial Technology Research Institute, Hsinchu, Taiwan, R.O.C ǂ
Stop Line Detection and Distance Measurement for Road Intersection based on Deep Learning Neural Network Guan-Ting Lin 1, Patrisia Sherryl Santoso *1, Che-Tsung Lin *ǂ, Chia-Chi Tsai and Jiun-In Guo National
More informationPatterns and Computer Vision
Patterns and Computer Vision Michael Anderson, Bryan Catanzaro, Jike Chong, Katya Gonina, Kurt Keutzer, Tim Mattson, Mark Murphy, David Sheffield, Bor-Yiing Su, Narayanan Sundaram and the rest of the PALLAS
More informationCONVOLUTIONAL COST AGGREGATION FOR ROBUST STEREO MATCHING. Somi Jeong Seungryong Kim Bumsub Ham Kwanghoon Sohn
CONVOLUTIONAL COST AGGREGATION FOR ROBUST STEREO MATCHING Somi Jeong Seungryong Kim Bumsub Ham Kwanghoon Sohn School of Electrical and Electronic Engineering, Yonsei University, Seoul, Korea E-mail: khsohn@yonsei.ac.kr
More informationHistograms of Sparse Codes for Object Detection
Histograms of Sparse Codes for Object Detection Xiaofeng Ren (Amazon), Deva Ramanan (UC Irvine) Presented by Hossein Azizpour What does the paper do? (learning) a new representation local histograms of
More informationCascade Region Regression for Robust Object Detection
Large Scale Visual Recognition Challenge 2015 (ILSVRC2015) Cascade Region Regression for Robust Object Detection Jiankang Deng, Shaoli Huang, Jing Yang, Hui Shuai, Zhengbo Yu, Zongguang Lu, Qiang Ma, Yali
More informationSSD: Single Shot MultiBox Detector. Author: Wei Liu et al. Presenter: Siyu Jiang
SSD: Single Shot MultiBox Detector Author: Wei Liu et al. Presenter: Siyu Jiang Outline 1. Motivations 2. Contributions 3. Methodology 4. Experiments 5. Conclusions 6. Extensions Motivation Motivation
More informationConvolutional-Recursive Deep Learning for 3D Object Classification
Convolutional-Recursive Deep Learning for 3D Object Classification Richard Socher, Brody Huval, Bharath Bhat, Christopher D. Manning, Andrew Y. Ng NIPS 2012 Iro Armeni, Manik Dhar Motivation Hand-designed
More informationDetecting Faces Using Inside Cascaded Contextual CNN
Detecting Faces Using Inside Cascaded Contextual CNN Kaipeng Zhang 1, Zhanpeng Zhang 2, Hao Wang 1, Zhifeng Li 1, Yu Qiao 3, Wei Liu 1 1 Tencent AI Lab 2 SenseTime Group Limited 3 Guangdong Provincial
More informationR-FCN++: Towards Accurate Region-Based Fully Convolutional Networks for Object Detection
The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18) R-FCN++: Towards Accurate Region-Based Fully Convolutional Networks for Object Detection Zeming Li, 1 Yilun Chen, 2 Gang Yu, 2 Yangdong
More informationPredicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture David Eigen, Rob Fergus
Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture David Eigen, Rob Fergus Presented by: Rex Ying and Charles Qi Input: A Single RGB Image Estimate
More informationDeep Learning in Visual Recognition. Thanks Da Zhang for the slides
Deep Learning in Visual Recognition Thanks Da Zhang for the slides Deep Learning is Everywhere 2 Roadmap Introduction Convolutional Neural Network Application Image Classification Object Detection Object
More informationLinear combinations of simple classifiers for the PASCAL challenge
Linear combinations of simple classifiers for the PASCAL challenge Nik A. Melchior and David Lee 16 721 Advanced Perception The Robotics Institute Carnegie Mellon University Email: melchior@cmu.edu, dlee1@andrew.cmu.edu
More informationMoFA: Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction
MoFA: Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction Ayush Tewari Michael Zollhofer Hyeongwoo Kim Pablo Garrido Florian Bernard Patrick Perez Christian Theobalt
More informationImageNet Classification with Deep Convolutional Neural Networks
ImageNet Classification with Deep Convolutional Neural Networks Alex Krizhevsky Ilya Sutskever Geoffrey Hinton University of Toronto Canada Paper with same name to appear in NIPS 2012 Main idea Architecture
More informationStructured Forests for Fast Edge Detection
2013 IEEE International Conference on Computer Vision Structured Forests for Fast Edge Detection Piotr Dollár Microsoft Research pdollar@microsoft.com C. Lawrence Zitnick Microsoft Research larryz@microsoft.com
More informationarxiv: v3 [cs.cv] 21 Sep 2015
High-for-Low and Low-for-High: Efficient Boundary Detection from Deep Object Features and its Applications to High-Level Vision arxiv:1504.06201v3 [cs.cv] 21 Sep 2015 Gedas Bertasius University of Pennsylvania
More informationDeep Learning for Computer Vision II
IIIT Hyderabad Deep Learning for Computer Vision II C. V. Jawahar Paradigm Shift Feature Extraction (SIFT, HoG, ) Part Models / Encoding Classifier Sparrow Feature Learning Classifier Sparrow L 1 L 2 L
More informationReal-time Object Detection CS 229 Course Project
Real-time Object Detection CS 229 Course Project Zibo Gong 1, Tianchang He 1, and Ziyi Yang 1 1 Department of Electrical Engineering, Stanford University December 17, 2016 Abstract Objection detection
More informationDirect Multi-Scale Dual-Stream Network for Pedestrian Detection Sang-Il Jung and Ki-Sang Hong Image Information Processing Lab.
[ICIP 2017] Direct Multi-Scale Dual-Stream Network for Pedestrian Detection Sang-Il Jung and Ki-Sang Hong Image Information Processing Lab., POSTECH Pedestrian Detection Goal To draw bounding boxes that
More informationImage Captioning with Object Detection and Localization
Image Captioning with Object Detection and Localization Zhongliang Yang, Yu-Jin Zhang, Sadaqat ur Rehman, Yongfeng Huang, Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
More informationDeep Convolutional Neural Networks. Nov. 20th, 2015 Bruce Draper
Deep Convolutional Neural Networks Nov. 20th, 2015 Bruce Draper Background: Fully-connected single layer neural networks Feed-forward classification Trained through back-propagation Example Computer Vision
More informationSpatial Localization and Detection. Lecture 8-1
Lecture 8: Spatial Localization and Detection Lecture 8-1 Administrative - Project Proposals were due on Saturday Homework 2 due Friday 2/5 Homework 1 grades out this week Midterm will be in-class on Wednesday
More informationTA Section: Problem Set 4
TA Section: Problem Set 4 Outline Discriminative vs. Generative Classifiers Image representation and recognition models Bag of Words Model Part-based Model Constellation Model Pictorial Structures Model
More informationarxiv: v2 [cs.cv] 23 May 2016
Localizing by Describing: Attribute-Guided Attention Localization for Fine-Grained Recognition arxiv:1605.06217v2 [cs.cv] 23 May 2016 Xiao Liu Jiang Wang Shilei Wen Errui Ding Yuanqing Lin Baidu Research
More informationCS 1674: Intro to Computer Vision. Neural Networks. Prof. Adriana Kovashka University of Pittsburgh November 16, 2016
CS 1674: Intro to Computer Vision Neural Networks Prof. Adriana Kovashka University of Pittsburgh November 16, 2016 Announcements Please watch the videos I sent you, if you haven t yet (that s your reading)
More informationPixels to Voxels: Modeling Visual Representation in the Human Brain
Pixels to Voxels: Modeling Visual Representation in the Human Brain Authors: Pulkit Agrawal, Dustin Stansbury, Jitendra Malik, Jack L. Gallant Presenters: JunYoung Gwak, Kuan Fang Outlines Background Motivation
More informationGEDContours: A High Speed Contour Detector for Grayscale Images
GEDContours: A High Speed Contour Detector for Grascale Images Cunet Akinlar cakinlar@anadolu.edu.tr Anadolu Universit, Computer Engineering Department, Eskisehir, Turke Abstract We propose a high-speed
More informationROBUST INTERNAL EXEMPLAR-BASED IMAGE ENHANCEMENT. Yang Xian 1 and Yingli Tian 1,2
ROBUST INTERNAL EXEMPLAR-BASED IMAGE ENHANCEMENT Yang Xian 1 and Yingli Tian 1,2 1 The Graduate Center, 2 The City College, The City University of New York, New York, Email: yxian@gc.cuny.edu; ytian@ccny.cuny.edu
More informationTowards Real-Time Automatic Number Plate. Detection: Dots in the Search Space
Towards Real-Time Automatic Number Plate Detection: Dots in the Search Space Chi Zhang Department of Computer Science and Technology, Zhejiang University wellyzhangc@zju.edu.cn Abstract Automatic Number
More informationConvolutional Neural Networks: Applications and a short timeline. 7th Deep Learning Meetup Kornel Kis Vienna,
Convolutional Neural Networks: Applications and a short timeline 7th Deep Learning Meetup Kornel Kis Vienna, 1.12.2016. Introduction Currently a master student Master thesis at BME SmartLab Started deep
More informationYiqi Yan. May 10, 2017
Yiqi Yan May 10, 2017 P a r t I F u n d a m e n t a l B a c k g r o u n d s Convolution Single Filter Multiple Filters 3 Convolution: case study, 2 filters 4 Convolution: receptive field receptive field
More informationMultiple Kernel Learning for Emotion Recognition in the Wild
Multiple Kernel Learning for Emotion Recognition in the Wild Karan Sikka, Karmen Dykstra, Suchitra Sathyanarayana, Gwen Littlewort and Marian S. Bartlett Machine Perception Laboratory UCSD EmotiW Challenge,
More informationFACIAL POINT DETECTION USING CONVOLUTIONAL NEURAL NETWORK TRANSFERRED FROM A HETEROGENEOUS TASK
FACIAL POINT DETECTION USING CONVOLUTIONAL NEURAL NETWORK TRANSFERRED FROM A HETEROGENEOUS TASK Takayoshi Yamashita* Taro Watasue** Yuji Yamauchi* Hironobu Fujiyoshi* *Chubu University, **Tome R&D 1200,
More informationPartial Least Squares Regression on Grassmannian Manifold for Emotion Recognition
Emotion Recognition In The Wild Challenge and Workshop (EmotiW 2013) Partial Least Squares Regression on Grassmannian Manifold for Emotion Recognition Mengyi Liu, Ruiping Wang, Zhiwu Huang, Shiguang Shan,
More informationContext by Region Ancestry
Context by Region Ancestry Joseph J. Lim, Pablo Arbeláez, Chunhui Gu, and Jitendra Malik University of California, Berkeley - Berkeley, CA 94720 {lim,arbelaez,chunhui,malik}@eecs.berkeley.edu Abstract
More informationAK Computer Vision Feature Point Detectors and Descriptors
AK Computer Vision Feature Point Detectors and Descriptors 1 Feature Point Detectors and Descriptors: Motivation 2 Step 1: Detect local features should be invariant to scale and rotation, or perspective
More informationWavelet-based Texture Classification of Tissues in Computed Tomography
Wavelet-based Texture Classification of Tissues in Computed Tomography Lindsay Semler, Lucia Dettori, Jacob Furst Intelligent Multimedia Processing Laboratory School of Computer Science, Telecommunications,
More information