
Computer Vision
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
The two volumes CCIS 546 and 547 constitute the refereed proceedings of the CCF Chinese Conference on Computer Vision, CCCV 2015, held in Xi'an, China, in September 2015.
The total of 89 revised full papers presented in both volumes were carefully reviewed and selected from 176 submissions. The papers address issues such as computer vision, machine learning, pattern recognition, target recognition, object detection, target tracking, image segmentation, image restoration, face recognition, image classification.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Organizers
- Contents - Part I
- Contents - Part II
- Fusion Technique for Infrared and Visible Images Based on Improved Quantum Theory Model
- 1 Introduction
- 2 Traditional Quantum Theory Model
- 3 Proposed Fusion Technique
- 3.1 Improved Quantum Theory Model
- 3.2 Concrete Fusion Algorithm
- 4 Experimental R Results and Discussion
- 5 Conclusions
- References
- Skeleton-Based Human Action Recognition with Profile Hidden Markov Models
- 1 Introduction
- 2 Related Work
- 3 Proposed Method
- 3.1 Representation of Meaningful Action Units
- 3.2 Clustering Feature Using Unsupervised Learning
- 3.3 Profile HMMs for Temporal Alignment of Human Motion
- 4 Experimental Evaluation
- 4.1 Evaluation Settings
- 4.2 Experimental Results
- 5 Conclusions and Future Work
- References
- Single Image Super-Resolution via Blind Blurring Estimation and Dictionary Learning
- 1 Introduction
- 2 Related Work
- 2.1 Internal Statistics and Blur Kernel Estimation
- 2.2 Coupled Dictionary Learning
- 3 Proposed Approach
- 3.1 Blind Kernel Estimation with SPP
- 3.2 SR Recovery
- 4 Experimental Results
- 4.1 Comparisons for BKE
- 5 Conclusion
- References
- Two-layers Local Coordinate Coding
- 1 Introduction
- 2 Prior Art
- 3 Two-Layers Local Coordinate Coding
- 3.1 Lipschitz Smoothness
- 3.2 Two-layers Coordinate Coding
- 4 Two-Layers Coding Formulation
- 4.1 First-Layer Formulation
- 4.2 Second-Layer Formulation
- 5 Experiment Verification
- 5.1 Quantitative Results
- 5.2 Applications in Robotics
- 6 Conclusions
- References
- A Graph Matching and Energy Minimization Based Algorithm for Lunar Surface Image Mosaic
- 1 Introduction
- 2 The Proposed Lunar Surface Image Mosaic Algorithm
- 2.1 Graph Matching Based Image Registration
- Feature Extraction.
- Feature Correspondence.
- Image Transformation and Interpolation.
- 2.2 Energy Minimization Based Image Stitching
- Exposure Compensation.
- Pixel Labelling.
- Gradient Reconstruction.
- 3 Experiment
- 4 Conclusion
- References
- A Real Time Robust Hand Tracking Method with Normal Cameras
- 1 Introduction
- 2 Efficient Tracking Framework
- 2.1 Initial Hand Detection
- 2.2 Segmentation Module
- 2.3 Tracking Module
- 3 Enhancements for Robust Tracking
- 3.1 Failure Compensation
- 3.2 Adaptive Threshold Segmentation
- 3.3 Interference Alert/Mitigation
- 4 Tracking Performance
- 4.1 Evaluation Setup
- 4.2 Tracking Efficiency
- 4.3 Tracking Robustness
- 5 Conclusions
- References
- Hypergraph Regularized Autoencoder for 3D Human Pose Recovery
- 1 Introduction
- 2 Hypergraph Regularized Autoencoder
- 2.1 Marginalized Denoising Autoencoders
- 2.2 Manifold Regularization
- 2.3 Hypergraph Optimization
- 3 Experimental Evaluation
- 3.1 Datasets and Settings
- 3.2 Optimization of Autoencoders
- 3.3 Comparison with State-of-the-Arts
- 4 Conclusion
- References
- Target Recognition Based on Feature Weighted Intuitionistic FCM
- 1 Introduction
- 2 Algorithm for Feature Weighted Intuitionistic Fuzzy
- means Clustering
- 3 Experimental Results and Analysis
- 3.1 FWIFCM Algorithm's Clustering Experiment
- 3.2 Target Recognition Simulation Instance Based on the Algorithm of FWIFCM
- 4 Conclusions
- References
- The Modified Mumford-Shah Model Based on Nonlocal Means Method for Textures Segmentation
- 1 Introduction
- 2 The Mumford-Shah Model
- 3 The Modified Mumford-Shah Model Based on Nonlocal Means Method
- 3.1 Nonlocal Operators
- 3.2 The Proposed Model
- 4 Experiment and Analysis
- 5 Conclusions
- References
- Segmentation-Based Salient Object Detection
- 1 Introduction
- 2 Method
- 2.1 Salient Object Detection with Segmentation
- 2.2 A New Dataset Labeled with Hierarchical Salient Objects
- 3 Experimental Results
- 4 Conclusion and Discussion
- References
- Non-uniform Motion Deblurring Using Normalized Hyper Laplacian Prior
- 1 Introduction
- 2 Related Work
- 3 Proposed Method
- 3.1 Framework of Efficient Filter Flow
- 3.2 Normalized Hyper Laplacian Prior in EFF
- 3.3 Removing Poorly Estimated Blur Kernels
- 3.4 Re-estimating Erroneous Local Kernels
- 4 Numerical Experiments
- 5 Conclusion
- References
- Robust Mean Shift Tracking Based on Refined Appearance Model and Online Update
- 1 Introduction
- 2 Mean Shift Tracking Algorithm
- 3 The Proposed Tracking Algorithm
- 3.1 Simple Tracking
- 3.2 Model Construction
- 3.3 Tracking with Refined Model
- 3.4 Model Update
- 3.5 Summary of the Whole Algorithm
- 4 Experiments
- 5 Discussion
- 6 Conclusion
- References
- An Effective Multiview Stereo Method for Uncalibrated Images
- 1 Introduction
- 2 Overview
- 3 Match Expansion
- 4 Point-Cloud Expansion
- 4.1 Optimization
- 4.2 Expansion
- 5 Experimental Results and Comparisons
- 6 Conclusion
- References
- The Human Image Segmentation Algorithm Based on Face Detection and Biased Normalized Cuts
- 1 Introduction
- 2 Related Work
- 2.1 Face Detection
- 2.2 Human Image Segmentation
- 3 Our Algorithm Based on Face Detection and Biased Normalized Cuts
- 4 Analysis and Experiments
- 5 Conclusions and Future Work
- References
- Survey on Eye Movement Based Authentication Systems
- 1 Introduction
- 2 Eye Movement Based Authentication Framework and Technical Modules
- 3 Cases Study
- 3.1 FBI + APM + KBID
- 3.2 FBI + GPM + KBID
- 3.3 FBI + GPM + BBID
- 3.4 SBI + APM + KBID
- 3.5 SBI + GPM + BBID
- 4 Discussion
- 4.1 Fixation vs. Saccade
- 4.2 APM vs. GPM and KBID vs. BBID
- 4.3 Summary and Comparison of State-of-the-Art EMBA Systems Technical Details
- 5 Conclusion
- References
- A Discriminant Sparse Representation Graph-Based Semi-Supervised Learning for Hyperspectral Image Classification
- 1 Introduction
- 2 Related Works
- 3 Discriminant Sparse Representation Graph Construction
- 3.1 Estimation of Class-Probability
- 3.2 Discriminant Sparse Representation Graph
- 4 Experiments and Analysis
- 4.1 Experimental Datasets
- 4.2 Results of Classification Experiments
- 5 Conclusion
- References
- Multispectral Image Classification Using a New Bayesian Approach with Weighted Markov Random Fields
- 1 Introduction
- 2 The Proposed Method
- 2.1 GMM Modeling for Likelihood Function
- 2.2 Anisotropic MLL for Spatial Prior
- 2.3 Computing the MAP Estimate via Simulated Annealing
- 3 Experiments
- 3.1 Data Sets
- 3.2 Experimental Results
- 4 Conclusion
- References
- Effective Facial Expression Recognition via the Boosted Convolutional Neural Network
- 1 Introduction
- 2 Related Works
- 3 The Proposed Boosted-CNN Method
- 3.1 Overview
- 3.2 Reducing Over-Fitting
- 3.3 Boosting Algorithm
- 4 Experimental Results
- 4.1 Experiment on the CK+ Database
- 4.2 Experiments on JAFFE Database
- 5 Conclusion and Future Work
- References
- Structured Sparse Coding for Classification via Reweighted 2,1 Minimization
- 1 Introduction
- 2 Related Work
- 3 Our Method
- 3.1 Problem Formulation
- 3.2 The Proposed Iterative Algorithm
- 3.3 Classification Process
- 4 Experiment
- 4.1 Implementation Details
- 4.2 Face Recognition
- 4.3 Gender Classification
- 4.4 Scene Classification
- 5 Summary
- References
- Detail-Enhanced Cross-Modality Face Synthesis via Guided Image Filtering
- 1 Introduction
- 2 Related Work
- 3 Cross-Modality Face Synthesis
- 3.1 K-NN-Based Algor rithm
- 3.2 Guided Image Filtering-Based Algorithm
- 4 Experimental Results
- 4.1 Evaluation on the CUHK Benchmark Dataset
- 4.2 Visible-to-Near-Infrared Face Synthesis
- 5 Conclusions
- References
- Edge-Directed Single Image Super-Resolution via Cross-Resolution Sharpening Function Learning
- 1 Introduction
- 2 Edge-Directed Single-Image Super-Resolution
- 3 Cross-Resolution Gradient Magnitude Sharpening
- 3.1 Construction of Tl
- 3.2 High-Resolution Gradient Construction
- 4 Implementation
- 5 Experiments and Analysis
- 5.1 Comparisons with the State-of-the-Art Approaches
- 5.2 Large-Scale Comparison
- 5.3 Comparisons with Edge-Directed Reconstruction Method
- 6 Conclusion and Discussion
- References
- Compressive Tracking Based on Particle Filter
- 1 Introduction
- 2 Fast Compressive Tracking
- 3 Proposed Method
- 3.1 The Element Number of Measurement Vector
- 3.2 Calculation Method of Particle Weights
- 4 Experiments
- 4.1 Quantitative Evaluation
- 4.2 Visual Evaluation
- 5 Conclusions
- References
- An Adaptive Threshold Algorithm for Moving Object Segmentation
- 1 Introduction
- 2 Object Segmentation Based on Fixed Threshold
- 3 Object Segmentation Algorithm Based on Adaptive Threshold
- 3.1 The Basic Steps of the Algorithm
- 3.2 Threshold Initialization Based on Gaussian Mixture Model and Grabcut
- 3.3 Threshold Learning Based on K-Means Clustering
- 3.4 Threshold Updating
- 4 Experimental Results
- 5 Conclusions
- References
- Interactive Registration and Segmentation for Multi-Atlas-Based Labeling of Brain MR Image
- 1 Introduction
- 2 Method
- 2.1 Image Registration via Minimum Spanning Tree
- 2.2 Segmentation via Label Fusion
- 2.3 Interactive Registration and Segmentation
- 2.4 Summary: The Interactive Registration-Segmentation Pipeline
- 3 Experimental Results
- 4 Discussion
- Negative-Supervised Cascaded Deep Learning for Traffic Sign Classification
- 1 Introduction
- 2 Related Work
- 3 Our Method
- 3.1 Deep Convolutional Neural Networks
- 3.2 Negative-Supervised Learning
- 3.3 Traffic Sign Classification
- 4 Experiments
- 4.1 Database and Negative Library
- 4.2 Training Convolutional Neural Networks
- 4.3 Experimental Results
- 5 Conclusion and Discussion
- References
- Saliency Detection Based on Non-uniform Quantification for RGB Channels and Weights for Lab Channels
- 1 Introduction
- 2 Non-uniform Quantification for RGB Channels
- 2.1 Visual Sensitivity Tests on RGB Channels
- 2.2 Theory Basis of the Visual Sensitivity Order
- 2.3 Non-uniform Quantification for RGB Channels
- 3 Computing the Weights for Lab Channels
- 4 Saliency Detection Algorithm Based on Non-uniform Quantification and Channel Weights
- 5 Experimental Results and Analysis
- 6 Conclusion
- References
- Edge-Based Unsupervised Evaluation of Image Segmentation
- 1 Introduction
- 2 Edge-Based Segmentation Evaluation
- 2.1 Symbols Definition
- 2.2 Related Work
- Color-Based Segmentation Evaluation
- Entropy-Based Segmentation Evaluation
- 2.3 The Edge-Based Evaluation Criteria
- 2.4 Edge-Based Evaluation Function
- 3 Experiments
- 3.1 Performance of Segmentation Evaluation
- 3.2 Region Evaluation Performance
- 4 Conclusions
- References
- An Adaptive Fuzzy Clustering Algorithm Based on Multi-threshold for Infrared Image Segmentation
- 1 Introduction
- 2 Adaptive Fuzzy Clustering Algorithm Based on Multi-threshold for Infrared Image Segmentation
- 2.1 Multi-threshold Pseudo-Peak Removal
- 2.2 Adaptive Fuzzy Clustering
- 2.3 Algorithm
- 3 Experimental Results and Analysis
- 4 Conclusions
- References
- Locally Linear Embedding Based Dynamic Texture Synthesis
- 1 Introduction
- 2 Appearance Model
- 3 Dynamic Model
- 4 Observation Model
- 5 Experiments
- 6 Conclusions
- References
- A Deep Joint Learning Approach for Age Invariant Face Verification
- 1 Introduction
- 2 Related Work
- 3 Deep Joint Metric Learning Framework
- 3.1 Optimization Objective
- 3.2 Deep Architecture
- 3.3 Learning Algorithm
- 4 Experiment
- 5 Conclusion
- References
- A New Similarity Measure for Non-local Means Denoising
- 1 Introduction
- 2 NLM Algorithm
- 3 Improved NLM Algorithm
- 3.1 Improved LBP Operator
- 3.2 Rotation-Invariant and Noise-Resistant Similarity Measurement
- 3.3 Automatic Selection of Similar Sets
- 4 Experiments and Discussion
- 5 Conclusion
- References
- Mixture Models for Object Detection
- 1 Introduction
- 2 Multiple Components Generation
- 3 Discovering Mid-level Part Models
- 3.1 Clustering
- 3.2 Training
- 3.3 Part Selecting
- 4 Experiments
- 5 Conclusion
- References
- An Image Forensic Technique Based on 2D Lighting Estimation Using Spherical Harmonic Frames
- 1 Introduction
- 2 Lighting Model Using Bases and Frames
- 3 Lighting Estimation
- 3.1 2D Information Extraction
- 4 Experiments
- 4.1 Synthetic Data
- 4.2 Real Images
- 4.3 Tampered Images
- 5 Conclusion
- References
- Class Relatedness Oriented Discriminative Dictionary Learning
- 1 Introduction
- 2 A Brief Review of the DDL Models
- 3 Main Results
- 3.1 Model Training
- 3.2 Model Testing
- 4 Experiments
- 4.1 Object Classification
- 4.2 Face Recognition
- 5 Conclusions
- References
- Building an Effective Template Dictionary for Robust Offline Video Tracking
- 1 Introduction
- 2 Sparse Representation for Visual Tracking
- 3 Proposed Tracking System
- 3.1 Template Dictionary Construction
- Non-polluted Templates.
- Variational Templates.
- 3.2 Two-Stage Sparse Optimization
- 3.3 Bi-tracking Process
- 3.4 Template Updating
- 4 Experimental Results
- 4.1 Different Dictionaries
- 4.2 Different Tracking Processes
- 4.3 Comparison with Other Tracking Algorithms
- 5 Conclusion
- References
- Learning Blur Invariant Face Descriptors for Face Verification Under Realistic Environment
- 1 Introduction
- 2 Blur Invariant Face Descriptor via Multiple Metric Learning
- 2.1 Proposed Framework
- 2.2 Multi-scale Absolute Phase Difference Feature
- 2.3 Fisher Kernel Encoding
- 2.4 Multiple Metric Learning
- 3 Experiment
- 3.1 Blurred FERET Face Dataset
- 3.2 Realistic Face Dataset
- 3.3 Parameter Analysis
- 3.4 Computational Efficiency
- 4 Conclusion
- References
- Modified Numerical Scheme for Perona-Malik Model in Image Restoration
- 1 Introduction
- 2 Nonlinear Diffusion in Image Restoration
- 2.1 Perona-Malik Model
- 2.2 Modified Numerical Scheme
- 2.3 Compatibility, Stability and Convergence
- 3 Numerical Comparison
- 4 Conclusion
- References
- Abnormal Event Detection Based on Multi-scale Markov Random Field
- 1 Introduction
- 2 Activity Representation
- 2.1 Features and the Corresponding Similarity
- 2.2 Model Acquisition and Maintenance
- 3 Abnormality Detection Based on Multi-scale MRF
- 3.1 Structure of Multi-scale MRF
- 3.2 Energy Function of the Multi-scale Markov Random Field
- 4 Experimental Results
- 4.1 Global Abnormal Event Detection
- 4.2 Subway Station Dataset
- 5 Conclusion
- References
- Light-Weight Spatial Distribution Embedding of Adjacent Features for Image Search
- 1 Introduction
- 2 Related Work
- 3 The Proposed Approach
- 3.1 Large-Scale Visual Dictionary Construction
- 3.2 Light-Weight Binary Code Embedding
- 3.3 Indexing and Retrieval
- 4 Experiments
- 4.1 Datasets
- 4.2 Evaluation on Image Search Performance
- 5 Conclusion
- References
- Cross-Level: A Practical Strategy for Convolutional Neural Networks Based Image Classification
- 1 Introduction
- 2 Related Work
- 3 Cross-Level
- 4 Experiments
- 5 Conclusion
- References
- Superpixel-Based Global Contrast Driven Saliency Detection in Low Contrast Images
- 1 Introduction
- 2 Proposed Algorithm
- 2.1 Superpixel Segmentation
- 2.2 Global Contrast Approach
- 2.3 Internal Similarity Measure
- 3 Experimental Results
- 4 Conclusions
- References
- Decoupled Marginal Distribution of Gradient Magnitude and Laplacian of Gaussian for Texture Classification
- 1 Introduction
- 1.1 Related Work
- 1.2 Contributions
- 2 Texture Classification Based on GM and LOG
- 2.1 Marginal Distributions of GM and LOG
- 2.2 Normalization of GM and LOG
- 2.3 Texture Image Representation and Classification
- 3 Classification Results and Analysis
- 3.1 Texture Datasets and Experiments Settings
- 3.2 Effect of the Normalization and Patch Size
- 3.3 Performance Comparison on the Three Datasets
- 4 Conclusions
- References
- MR Image Segmentation Using Active Contour Model Incorporated with Sobel Edge Detection
- 1 Introduction
- 2 An Edge Detection Model Based on Sobel Operator
- 3 Region-Based Active Contour Model Combines with Sobel Operater
- 4 The Distance Regularized Level Set Formulation of the Function
- 5 Experimental Result
- 6 Conclusion
- Refeerences
- Spatial-Temporal Feature Fusion for Human Fall Detection
- 1 Introduction
- 2 The Proposed Approach
- 2.1 Preprocessing
- 2.2 Spatial-Temporal Features
- 2.3 Feature Encoding
- 2.4 Classification and Fusion
- 3 Experimental Results
- 3.1 Dataset
- 3.2 Settings
- 3.3 Results
- 4 Conclusions
- References
- Color Image Segmentation Combining Rough Depth Information
- 1 Introduction
- 2 Algorithm
- 2.1 Preprocess
- 2.2 Adaptive Mean Shift Image Segmentation
- 2.3 Probability Boundary with Depth Discontinuities
- 2.4 Graph Cut Based on GMM Modeling
- 3 Experiments
- 4 Conclusion
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.