Computer Vision - ACCV 2010
10th Asian Conference on Computer Vision, Queenstown, New Zealand, November 8-12, 2010, Revised Selected Papers, Part I
Published on 2. March 2011
XXI, 510 pages
978-3-642-19315-6 (ISBN)
Description
The four-volume set LNCS 6492-6495 constitutes the thoroughly refereed post-proceedings of the 10th Asian Conference on Computer Vision, ACCV 2009, held in Queenstown, New Zealand in November 2010.
All together the four volumes present 206 revised papers selected from a total of 739 Submissions. All current issues in computer vision are addressed ranging from algorithms that attempt to automatically understand the content of images, optical methods coupled with computational techniques that enhance and improve images,
and capturing and analyzing the world's geometry while preparing the higher level image and shape understanding. Novel gemometry techniques, statistical learning methods, and modern algebraic procedures are dealt with as well.
More details
Other editions
Additional editions
Ron Kimmel | Reinhard Klette | Akihiro Sugimoto
Computer Vision - ACCV 2010
10th Asian Conference on Computer Vision, Queenstown, New Zealand, November 8-12, 2010, Revised Selected Papers, Part I
Book
03/2011
Springer
€53.49
Shipment within 7-9 days
Content
- Title Page
- Preface
- Organization
- Table of Contents - Part I
- Keynote
- Deformable Object Modelling and Matching
- Statistical Shape Models
- Active Shape Models
- Active Appearance Models
- Automatic Model Construction
- Initialising Groupwise Registration
- Conclusions
- References
- Geometry and Correspondence
- New Efficient Solution to the Absolute Pose Problem for Camera with Unknown Focal Length and Radial Distortion
- Introduction
- Problem Formulation
- Absolute Pose for a Camera with Unknown Focal Length and Radial Distortion for Non-planar Scene
- Absolute Pose for a Camera with Unknown Focal Length and Radial Distortion for Planar Scene
- Experiments
- Synthetic Datasets
- Real Data
- Conclusion
- References
- Efficient Large-Scale Stereo Matching
- Introduction
- Related Work
- Efficient Large-Scale Stereo Matching
- Support Points
- Generative Model for Stereo Matching
- Disparity Estimation
- Experimental Evaluation
- Entropy Reduction
- Accuracy and Running Time for the Middlebury Dataset
- Urban Sequences and Face Image Set
- Conclusion and Future Work
- References
- Towards Full 3D Helmholtz Stereovision Algorithms
- Introduction
- Helmholtz Stereopsis - Variational Formulation
- Optimization for Triangle Mesh Representation
- Experimental Results
- Conclusion
- References
- Image-Based 3D Modeling via Cheeger Sets
- Introduction
- Variational Formulation
- Implicit Weighted Variational Surfaces
- Volume Constraint
- Fast Minimization
- Optimality Bounds
- Theoretical Analysis of Material Concentration
- Experimental Results
- Cheeger Sets and Single View Reconstruction
- Fixed Volume vs. Shape Prior
- Varying the Volume
- Weighted Minimal Surface Reconstruction
- Conclusion
- References
- Network Connectivity via Inference over Curvature-Regularizing Line Graphs
- Introduction
- Preliminaries
- Weighted Line Graph
- Graph Construction
- Minimum Energy Proposal Curves
- Expected Curve Energy
- Inferring Connectivity
- Experiments
- Simulated Data
- Anatomical Structures
- Conclusions
- References
- Computational Photography and Low Level Vision
- Image and Video Decolorization by Fusion
- Introduction
- Related Work
- Fusion-Based Decolorization Approach
- $H--K$ Chromatic Adapted Lightness
- Weight Maps Assignment
- Multi-scale Fusion of the Inputs
- Results and Discussion
- Evaluation of Grayscale Operators
- Video Decolorization
- Conclusion
- References
- Video Temporal Super-Resolution Based on Self-similarity
- Introduction
- Related Work
- Self-similarity
- Definition
- Necessary Conditions
- Temporal Super-Resolution Algorithm
- Overview
- Likelihood Model
- Prior Model
- Optimization
- Experimental Results
- Comparison with Other Methods
- Evaluation of Components
- Conclusions
- References
- Temporal Super Resolution from a Single Quasi-periodic Image Sequence Based on Phase Registration
- Introduction
- Related Work
- Problem Setting
- Assumptions
- Quasi-periodic Image Sequence
- Manifold Representation
- Energy Minimization Framework
- Energy Function
- Solution by Linear Approximation
- Dynamic Programming Solution
- Iterative Solution by Quadratic Approximation
- Experiments
- Experimental Setup
- Refinement Process
- Evaluation
- Discussion
- Conclusion
- References
- Solving MRFs with Higher-Order Smoothness Priors Using Hierarchical Gradient Nodes
- Introduction
- Preliminaries
- Composing Higher-Order Cliques Using Hierarchical Auxiliary Nodes
- Gradient Nodes: Reducing Complexity of Auxiliary Nodes
- Optimization
- Max-product BP on Factor Graphs and HGN Graphs
- TRW Message Passing on HGN Graphs
- Experimental Results
- Signal Reconstruction
- Image Denoising
- Sub-pixel Stereo Matching
- Nonrigid Image Registration
- Conclusion
- References
- An Efficient RANSAC for 3D Object Recognition in Noisy and Occluded Scenes
- Introduction
- Related Work
- Method Description
- Model Preprocessing Phase
- Online Recognition Phase
- Time Complexity
- Experimental Results
- Conclusions
- References
- Detection and Recognition I
- Change Detection for Temporal Texture in the Fourier Domain
- Introduction
- Statistical Model for Multiple Random Motions, Temporal Textures
- Statistical Model for One Type of Random Motion, Temporal Textures
- Change Detection for Temporal Textures
- Experiments
- Temporal Textures
- Random Motions in Groups
- Conclusions
- References
- Stream-Based Active Unusual Event Detection
- Introduction
- Related Work
- Active Unusual Event Detection
- Activity Representation
- Bayesian Classification
- Query Criteria
- Adaptive Selection of Multiple Query Criteria
- Experiments
- Datasets and Settings
- Active Learning vs. Unsupervised Learning
- Active Learning vs. Passive Supervised Learning and Other Active Learning Strategies
- Conclusion
- References
- Asymmetric Totally-Corrective Boosting for Real-Time Object Detection
- Introduction
- Asymmetric Losses
- Asymmetric Totally-Corrective Boosting
- Experiments
- Results on Synthetic Data
- Face Detection
- Conclusion
- References
- Keynote
- The Application of Vision Algorithms to Visual Effects Production
- Introduction
- The Use of Machine Vision Techniques in VFX
- Issues in Adapting Research Results
- Conclusions
- References
- Applied Computer Vision
- Automatic Workflow Monitoring in Industrial Environments
- Introduction
- Problem Formulation and Modelling
- Motion-Grid
- Echo State Network
- Workflow Monitoring
- Setup/Maintaince
- Run-Time
- Experimental Results
- Car Assembly Example Setup
- Implementation Details
- Evaluation and Results
- Conclusion
- References
- Context-Based Support Vector Machines for Interconnected Image Annotation
- Introduction
- Related Work
- Motivation and Contribution
- Kernel Design
- Context and Graph-Links
- Context-Dependent Kernel Design
- Solution
- Positive Definiteness
- Benchmarking
- Databases and Settings
- Hold-Out Generalization and Comparison
- Conclusion
- References
- Finding Human Poses in Videos Using Concurrent Matching and Segmentation
- Introduction
- Related Work
- Concurrent Matching and Segmentation
- Body Part Candidates and Superpixels
- The Linear Optimization
- Relaxation and Branch and Bound Solution
- Experimental Results
- Conclusion
- References
- Modeling Sense Disambiguation of Human Pose: Recognizing Action at a Distance by Key Poses
- Introduction
- Proposed Methodology
- Pose Descriptor: Combining Motion and Pose
- Unsupervised Learning of Visual Codebook of Poses
- What Does a Pose Tell Us? Modeling Sense Disambiguation of Human Pose
- Action Descriptor for Target Video Classification
- Experiments, Results and Discussions
- Experimental Setup
- Results and Discussions
- Conclusions and Future Scope
- References
- Social Interactive Human Video Synthesis
- Introduction
- Background
- Overview
- Human Video Texture Synthesis
- Video Data
- Identifying Cut Points
- Texture Motion Graph
- Dynamic Model
- Motion Synthesis
- Social Interaction Model
- Social Behaviour Experiment
- Semi-supervised Social Signal Labelling
- Data Mining for Social Trends
- Social Interactive Motion Control Using Apriori Mining
- Animation/Results
- Conclusion
- References
- Tracking and Categorization
- Efficient Visual Object Tracking with Online Nearest Neighbor Classifier
- Introduction
- Tracking with an Online Nearest-Neighbor Classifier
- Appearance Model
- Motion Model
- Algorithm and Implementation
- Analysis
- Experiments
- Failure Modes
- Conclusions and Future Work
- References
- Robust Tracking with Discriminative Ranking Lists
- Introduction
- Modeling a Pair of Object Models
- Locating the Object
- Updating Object Models
- Experiments
- Single-Scale Patch vs. Multi-scale Patch
- Ability to Remove Background Outliers
- Necessity of Purity
- Comparison with State-of-the-Art Approaches
- Conclusion
- References
- Analytical Dynamic Programming Tracker
- Introduction
- Related Work
- Conventional DP Tracker
- General Formulation of Tracking Problem
- Solution by Conventional DP
- Computational Complexity of Conventional DP
- The Analytical DP Tracker
- Quadratic Representation of Cost
- Solution by Analytical DP
- Global Optimality of Solution
- Computational Complexity
- Implementation Issues
- Quadratic Representation by Approximation
- Iterative Updating
- Experimental Results
- Tracking Results
- Computation Time
- Discussion
- Relation to KLT Tracker
- Bi-directional Strategy
- Multiple Quadratic Functions
- Conclusion
- References
- A Multi-Scale Learning Framework for Visual Categorization
- Introduction
- Related Work
- Multi-Scale Learning for Image Classification
- Experiments
- Datasets
- Dictionary Learning and Local Descriptor Encoding
- Training
- Results of the Oxford Flower Dataset
- Results of the Caltech-101 Dataset
- Conclusion
- References
- Image Sensing
- Fast Spectral Reflectance Recovery Using DLP Projector
- Introduction
- Related Work
- Spectral Reflectance Recovery
- Three Steps for Spectral Reflectance Recovery
- Image Acquisition by Color Switch
- Illumination Recovery
- Spectral Reflectance Reconstruction Using Constrained Model
- Accuracy Evaluation
- Image and Video Relighting
- Image Relighting
- Real Video Relighting
- Conclusion
- References
- Hemispherical Confocal Imaging Using Turtleback Reflector
- Introduction
- Related Work
- Reflectance Field Measurement
- BRDF Measurement Using Mirrors
- Descattering
- Hemispherical Confocal Imaging
- Optical Design of Turtleback Reflector
- Projectors and Cameras on a Hemisphere
- Turtleback Reflector for Virtual Projectors and Cameras
- Design of Optical Device
- Focused High Frequency Illumination
- Illumination and Reflection in a 3-D Scene
- Focused Illumination by Multiple Projectors
- Factorization of the Observed Views
- Experiments
- Synthetic Aperture Using Virtual Cameras
- Descattering by Focused High Frequency Illumination
- Factorization of the Observed Views
- Limitations
- Conclusion
- References
- Keynote
- Image-Based and Sketch-Based Modeling of Plants and Trees
- Introduction
- Image-Based Modeling of Plants
- Image-Based Modeling of Trees
- Sketch-Based Modeling of Trees
- Concluding Remarks
- References
- Segmentation and Texture
- MOMI-Cosegmentation: Simultaneous Segmentation of Multiple Objects among Multiple Images
- Introduction
- Related Work
- Common Pattern Discovery
- MOMI-Cosegmentation Incorporating Common Pattern Discovery
- Color Term and Smoothness Term
- Confidence Term
- Locality Term
- Experimental Results
- Conclusion
- References
- Spatiotemporal Closure
- Introduction
- Related Work
- Problem Formulation
- Algorithm Details
- Superpixel Extraction
- Superpixel Affinity
- Optimal Cuts for Each Shot
- Post-Processing
- Evaluation
- Conclusions
- References
- Compressed Sensing for Robust Texture Classification
- Introduction
- CS Background
- The CS Classifier
- The SCS Classifiers
- Sorted Pixel Values
- Sorted Pixel Differences
- Experimental Evaluation
- Image Data and Experimental Setup
- Experimental Results and Performance Analysis
- Conclusions
- References
- Interactive Multi-label Segmentation
- Introduction
- Method
- Features
- Model
- Multi-label Segmentation
- Experiments
- Benchmark
- Image Features
- Tooltip
- Random Forest / Repeatability
- Runtime
- Qualitative Examples
- Conclusion
- References
- Four Color Theorem for Fast Early Vision
- Introduction
- Loopy Belief Propagation
- Multiscale BP on the Grid Graph
- Forward Belief Propagation
- Four-Color Theorem in BP
- BP with Four-Colors
- FBP with Four-Colors
- Experiments
- Image Segmentation
- Image Restoration
- Stereo Matching
- Optical Flow
- Conclusions
- References
- Detection and Recognition II
- A Unified Approach to Segmentation and Categorization of Dynamic Textures
- Introduction
- Categorization of Videos of a Single Dynamic Texture Using a Bag of Dynamic Appearance Features
- Dynamic Texture Model
- Representation and Comparison of LDSs
- Classifiers
- Joint Segmentation and Categorization of Videos of Multiple Dynamic Textures
- Superpixels
- Classifiers
- Joint Segmentation and Categorization
- Experiments
- Whole Video Categorization
- Joint Segmentation and Categorization
- Conclusions
- References
- Learning Video Manifold for Segmenting Crowd Events and Abnormality Detection
- Introduction
- Related Work
- Crowd Video Analysis
- Manifold Learning in Video Content Analysis
- Proposed Method
- Representations of Video Frames
- Learning Video Trajectories: Using Spatio-temporal Laplacian Eigenmap (ST-LE)
- Clustering Video Trajectories: Similarity Measure
- Video Understanding Applications and Experimental Results
- Segmenting a Video Trajectory
- Abnormality Detection
- Conclusions
- References
- A Weak Structure Model for Regular Pattern Recognition Applied to Facade Images
- Introduction
- Structural Recognition Framework
- Structural Model
- Attribute Constraints
- Structural Prior
- Data Likelihood
- Edge Likelihood
- Color Likelihood
- Recognition Algorithm
- Metropolis-Hastings Moves
- Reversible Jump Moves
- Convergence and Complexity
- Experimental Results
- Conclusion and Future Work
- References
- Multiple Viewpoint Recognition and Localization
- Introduction
- Related Work
- Method
- Likelihood Model
- Object Detectors
- Identifying Potential 3D Objects
- Maximum Likelihood Object Localization
- Experiments
- Collecting Images from Multiple Registered Viewpoints
- Evaluation
- Conclusions
- References
- Matching and Similarity
- Localized Earth Mover's Distance for Robust Histogram Comparison
- Introduction
- Related Works
- Computation of LEMD and Lower Bounds
- Experimental Results
- Conclusion and Discussion
- References
- Geometry Aware Local Kernels for Object Recognition
- Introduction
- Geometry Aware Local Kernel
- Definition
- Special Cases
- Relation to Graph Kernels
- Mercer Condition
- Efficient Computation
- Performance
- Experiments and Results
- Discussion
- Conclusion
- References
- Author Index
