13032nam 22008175 450 99619826210331620200701105534.03-319-10590-610.1007/978-3-319-10590-1(CKB)3710000000219457(SSID)ssj0001338663(PQKBManifestationID)11743300(PQKBTitleCode)TC0001338663(PQKBWorkID)11345260(PQKB)11299532(DE-He213)978-3-319-10590-1(MiAaPQ)EBC5586520(Au-PeEL)EBL5586520(OCoLC)889238029(PPN)180626353(EXLCZ)99371000000021945720140813d2014 u| 0engurnn|008mamaatxtccrComputer Vision -- ECCV 2014[electronic resource] 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part I /edited by David Fleet, Tomas Pajdla, Bernt Schiele, Tinne Tuytelaars1st ed. 2014.Cham :Springer International Publishing :Imprint: Springer,2014.1 online resource (XXVIII, 853 p. 361 illus.) Image Processing, Computer Vision, Pattern Recognition, and Graphics ;8689Bibliographic Level Mode of Issuance: Monograph3-319-10589-2 Intro -- Foreword -- Preface -- Organization -- Table of Contents -- Tracking and Activity Recognition -- Visual Tracking by Sampling Tree-Structured Graphical Models -- 1 Introduction -- 2 Related Work -- 3 Main Framework -- 3.1 Learning Tree Structure by MCMC -- 3.2 Tracking on Tree Structure -- 4 Hierarchical Construction of Tree Structure -- 4.1 Tree Construction on Key Frames -- 4.2 Tree Extension by Manifold Alignment -- 5 Experiments -- 5.1 Datasets -- 5.2 Identified Tree Structure -- 5.3 Quantitative and Qualitative Performance -- 6 Conclusion -- References -- Tracking Interacting Objects Optimally Using Integer Programming -- 1 Introduction -- 2 Related Work -- 3 Method -- 3.1 Bayesian Inference -- 3.2 Flow Constraints -- 3.3 Mixed Integer Programming -- 3.4 Graph Size Reduction -- 4 Estimating Probabilities of Occupancy -- 4.1 Oriented Objects -- 4.2 Objects Off the Ground Plane -- 5 Experiments -- 5.1 Test Sequences -- 5.2 Parameters and Baselines -- 5.3 Results -- 6 Conclusion -- References -- Learning Latent Constituents for Recognition of Group Activities in Video -- 1 Introduction -- 2 Related Work -- 3 Learning Functional Constituents of Group Activities -- 3.1 Functional Grouping of Part Instances -- 3.2 From Constituent Classifiers to Activity Classification -- 3.3 Inference in Novel Query Scenes -- 3.4 Joint Learning of Group Activity and Constituent Classifiers -- 4 Experimental Results -- 4.1 Experimental Protocol -- 4.2 Group Activity Recognition -- 5 Conclusion -- References -- Recognition -- Large-Scale Object Classification Using Label Relation Graphs -- 1 Introduction -- 2 Related Work -- 3 Approach -- 3.1 Hierarchy and Exclusion (HEX) Graphs -- 3.2 Classification Model -- 3.3 Efficient Inference -- 4 Experiments -- 4.1 Implementation -- 4.2 Object Classification on ImageNet.4.3 Zero-Shot Recognition on Animals with Attributes -- 4.4 Efficiency of Inference -- 5 Discussions and Conclusions -- References -- 30Hz Object Detection with DPM V5 -- 1 Introduction -- 1.1 Prior Work -- 2 Pyramid of Features vs. Pyramid of Templates -- 3 Hierarchical Vector Quantization -- 4 Object Proposal Using Hash Table -- 4.1 Hash Codes -- 4.2 Priority Lists -- 4.3 Hash Table Initialization -- 5 Object Scoring -- 6 Experimental Results -- 7 Discussion -- References -- Knowing a Good HOG Filter When You See It: Efficient Selection of Filters for Detection -- 1 Introduction -- 1.1 Related Work -- 2 Background -- 2.1 An Overview of Poselets for Object Detection -- 2.2 An Overview of Exemplar SVMs for Object Detection -- 3 Ranking and Diversity -- 3.1 Learning to Rank Parts -- 3.2 Selecting a Diverse Set of Parts -- 3.3 Features for Part Ranking -- 3.4 The LDA Acceleration -- 4 Experiments with Poselets -- 4.1 Training the Ranking Algorithm -- 4.2 Training the Diversity Model -- 4.3 Selection Methods Considered -- 4.4 Ranking Results -- 4.5 PASCAL VOC Detection Results -- 5 Experiments with Exemplar SVMs -- 5.1 PASCAL VOC Detection Results -- 5.2 An Analysis of Bicycle HOG Filters -- 6 Conclusion -- References -- Linking People in Videos with "Their" Names Using Coreference Resolution -- 1 Introduction -- 2 Related Work -- 3 Problem Setup -- 4 Our Model -- 4.1 Regression-Based Clustering -- 4.2 Name Assignment to Tracks -- 4.3 Name Assignment to Mentions and Coreference Resolution -- 4.4 Alignment between Tracks and Mentions -- 5 Optimization -- 6 Experiments -- 6.1 Name Assignment to Tracks in Video -- 6.2 Name Assignment to Mentions -- 7 Conclusion -- References -- Poster Session 1 -- Optimal Essential Matrix Estimation via Inlier-Set Maximization -- 1 Introduction -- 1.1 Related Work -- 2 Essential Manifold Parametrization.3 Optimization Criteria -- 4 Branch and Bound over D2 π × B3 π -- 4.1 Lower-Bound Computation -- 4.2 Upper-Bound Computation via Relaxation -- 4.3 Efficient Bounding with Closed-Form Feasibility Test -- 4.4 The Main Algorithm -- 5 Experiment Results -- 5.1 Synthetic Scene Test: Normal Cases -- 5.2 Synthetic Scene Test: Special Cases -- 5.3 Real Image Test -- 6 Conclusion and Future Work -- References -- UPnP: An Optimal O(n) Solution to the Absolute Pose Problem with Universal Applicability -- 1 Introduction -- 1.1 Related Work -- 2 Theory -- 2.1 Geometry of the Absolute Pose Problem -- 2.2 Derivation of the Objective Function -- 2.3 Universal, Closed-Form Least-Squares Solution -- 2.4 Comparison to a Lagrangian Formulation -- 2.5 Elimination of Two-Fold Symmetry -- 2.6 Second-Order Optimality and Root Polishing -- 3 Experimental Evaluation -- 3.1 The Central Case -- 3.2 The Non-central Case -- 3.3 Computational Efficiency -- 3.4 Results on Real Data -- 4 Conclusion -- References -- 3D Reconstruction of Dynamic Textures in Crowd Sourced Data -- 1 Introduction -- 2 Related Work -- 3 Initial Model Generation -- 3.1 Static Reconstruction from Photo Collections -- 3.2 Coarse Dynamic Textures Priors from Video -- 3.3 Coarse Static Background Priors from Video Frames -- 3.4 Graph-Cut Based Dynamic Texture Refinement -- 3.5 Shape from Silhouettes -- 4 Closed Loop 3D Shape Refinement -- 4.1 Geometry Based Video to Image Label Transfer -- 4.2 Mitigating Dynamic Texture in SfM Estimates -- 4.3 Building a Static Background Prior for Single Images -- 4.4 Mitigating of Non-uniform Spatial Sampling -- 5 Experiments -- 6 Conclusion -- References -- 3D Interest Point Detection via Discriminative Learning -- 1 Introduction -- 2 Related Work -- 3 Attributes and Learning -- 3.1 Basic Attributes -- 3.2 DoG Attributes -- 3.3 Random Forest -- 3.4 Imbalanced Classes.4 Ground Truth and Experiments -- 4.1 Ground Truth -- 4.2 Experiments -- 4.3 Training and Predicting -- 4.4 Evaluation Criteria -- 4.5 Results -- 5 Conclusions -- References -- Pose Locality Constrained Representation for 3D Human Pose Reconstruction -- 1 Introduction -- 2 Related Work -- 3 Proposed Method -- 3.1 Problem Description -- 3.2 Hierarchical Pose Tree -- 3.3 Pose Locality for Reconstruction -- 3.4 Algorithm Summary -- 4 Experiments -- 4.1 Quantitative Results -- 4.2 Qualitative Evaluation -- 4.3 The Selection of Parameters dM and E -- 4.4 Distribution of Anchor-Node Depth -- 5 Conclusions -- References -- Synchronization of Two Independently Moving Cameras without Feature Correspondences -- 1 Introduction -- 1.1 Previous Work -- 1.2 Contributions -- 1.3 Notation and Paper Organization -- 2 Static and Jointly Moving Cameras -- 2.1 Video Synchronization -- 2.2 Uniqueness of Solution -- 3 Independently Moving Cameras -- 3.1 Video Synchronization -- 4 Object Motion Recovery -- 5 Experimental Results -- 6 Conclusions -- References -- Multi Focus Structured Light for Recovering Scene Shape and Global Illumination -- 1 Introduction -- 1.1 Related Work -- 2 Modeling Image Formation and Illumination -- 3 Illumination Control and Image Acquisition -- 4 Recovering Shape with Defocused Light Patterns -- 5 Recovering Direct and Global Illumination Components -- 6 Results -- 6.1 Depth Recovery -- 6.2 Recovering Direct and Global Illumination -- 7 Discussion -- References -- Coplanar Common Points in Non-centric Cameras -- 1 Introduction -- 2 Ray Space Analysis -- 2.1 CCPs in General Linear Cameras -- 2.2 Concentric Mosaics -- 3 CCPs in Catadioptric Mirrors -- 3.1 Ray Space vs. Caustics -- 3.2 Rotationally Symmetric Mirrors -- 3.3 Cylinder Mirrors -- 4 Experiments -- 4.1 Synthetic Experiment -- 4.2 Real Experiment -- 5 Conclusions and Discussions.References -- SRA: Fast Removal of General Multipath for ToF Sensors -- 1 Introduction -- 1.1 Contributions -- 2 Prior Work -- 3 Sparse Reflections Analysis -- 3.1 The Multipath Representation -- 3.2 The SRA Algorithm -- 4 Fast Computation -- 5 Experiments -- 5.1 Running Time -- 5.2 General Multipath: Examples -- 5.3 Comprehensive Two-Path Evaluation -- 5.4 Real Images -- 6 Conclusions -- References -- Sub-pixel Layout for Super-Resolution with Images in the Octic Group -- 1 Introduction -- 1.1 Contributions -- 1.2 Related Work -- 2 Good Sub-pixel Layout for Super-Resolution -- 2.1 Single Image Case -- 2.2 Multiple Images in the Octic Group -- 2.3 Good Sub-pixel Layout -- 3 Reconstruction Algorithm -- 4 Performance Evaluation -- 4.1 Synthetic Test -- 4.2 Real Data Test -- 5 Discussion -- 6 Conclusion -- References -- Simultaneous Feature and Dictionary Learning for Image Set Based Face Recognition -- 1 Introduction -- 2 Related Work -- 3 Proposed Approach -- 3.1 SFDL -- 3.2 Identification -- 3.3 Verification -- 4 Experimental Results -- 4.1 Datasets -- 4.2 Experimental Settings -- 4.3 Results and Analysis -- 5 Conclusion and Future Work -- References -- Read My Lips: Continuous Signer Independent Weakly Supervised Viseme Recognition -- 1 Introduction -- 2 Mouthings in Sign Language, Challenging? -- 3 State of the Art -- 4 Corpora -- 5 Mouthing Features -- 6 Weakly Supervised Mouthing Recognition -- 6.1 Overview -- 6.2 Reordering Sentence Structure -- 6.3 Pronunciation Lexicon and Viseme Mapping -- 6.4 Training Viseme Models -- 6.5 Context-Dependent Visemes with a Visemic Classification and Regression Tree -- 6.6 Linear Discriminant Analysis -- 7 Results -- 8 Conclusions -- References -- Multilinear Wavelets: A Statistical Shape Space for Human Faces -- 1 Introduction -- 2 Related Work -- 3 Multilinear Wavelet Model.3.1 Second Generation Spherical Wavelets.The seven-volume set comprising LNCS volumes 8689-8695 constitutes the refereed proceedings of the 13th European Conference on Computer Vision, ECCV 2014, held in Zurich, Switzerland, in September 2014. The 363 revised papers presented were carefully reviewed and selected from 1444 submissions. The papers are organized in topical sections on tracking and activity recognition; recognition; learning and inference; structure from motion and feature matching; computational photography and low-level vision; vision; segmentation and saliency; context and 3D scenes; motion and 3D scene analysis; and poster sessions.Image Processing, Computer Vision, Pattern Recognition, and Graphics ;8689Optical data processingPattern recognitionArtificial intelligenceComputer graphicsImage Processing and Computer Visionhttps://scigraph.springernature.com/ontologies/product-market-codes/I22021Pattern Recognitionhttps://scigraph.springernature.com/ontologies/product-market-codes/I2203XArtificial Intelligencehttps://scigraph.springernature.com/ontologies/product-market-codes/I21000Computer Graphicshttps://scigraph.springernature.com/ontologies/product-market-codes/I22013Optical data processing.Pattern recognition.Artificial intelligence.Computer graphics.Image Processing and Computer Vision.Pattern Recognition.Artificial Intelligence.Computer Graphics.006.6006.37Fleet Davidedthttp://id.loc.gov/vocabulary/relators/edtPajdla Tomasedthttp://id.loc.gov/vocabulary/relators/edtSchiele Berntedthttp://id.loc.gov/vocabulary/relators/edtTuytelaars Tinneedthttp://id.loc.gov/vocabulary/relators/edtMiAaPQMiAaPQMiAaPQBOOK996198262103316Computer Vision -- ECCV 20142587608UNISA