
Image and Video Technology
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
This
book constitutes the thoroughly refereed post-conference proceedings of the 7th
Pacific Rim Symposium on Image and Video Technology, PSIVT 2015, held in Auckland,
New Zealand, in November 2015.
The total of 61 revised papers was carefully reviewed and selected from 133 submissions. The papers are organized in topical sections on color and motion, image/video coding and transmission, computational photography and arts, computer vision and applications, image segmentation and classification, video surveillance, biomedical image processing and analysis, object and pattern recognition, computer vision and pattern recognition, image/video processing and analysis, and pattern recognition.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Organization
- Contents
- Color and Motion
- Color Conversion for Color Blindness Employing Multilayer Neural Network with Perceptual Model
- 1 Introduction
- 2 Proposed Method
- 2.1 Neural Networks
- 2.2 Color Model for Colorblindness
- 2.3 Neural Network Model for Simulating Colorblindness and Extraction of Color Edge
- 2.4 Algorithm of Proposed Method
- 3 Experimental Results and Discussions
- 4 Conclusions
- References
- Synthesis of Oil-Style Paintings
- 1 Introduction
- 2 Program Framework
- 3 Complexity Map and Point Map
- 4 Orientation Map
- 5 Stroke Design and Painting
- 5.1 Background Region
- 5.2 Edge Region
- 6 Experiment Results
- 7 Conclusion
- References
- Multi-frame Feature Integration for Multi-camera Visual Odometry
- 1 Introduction
- 2 Feature-Based Visual Odometry
- 3 Multi-frame Feature Integration
- 4 Multi-camera Multi-frame Feature Integration
- 4.1 Spatial-Temporal Feature Matching
- 4.2 Ego-Motion Estimation
- 4.3 Feature Integration and State Update
- 4.4 Lost Feature Recovery
- 5 Experimental Results
- 6 Conclusions
- References
- A Robust Identification Scheme for JPEG XR Images with Various Compression Ratios
- 1 Introduction
- 2 Background
- 2.1 Image Identification Model
- 2.2 Applications
- 2.3 JPEG XR
- 3 Proposed Identification Scheme
- 3.1 Notation and Terminologies
- 3.2 Identification Scheme
- 4 Simulation
- 4.1 Simulation Conditions
- 4.2 Evaluation for Still Images
- 4.3 Evaluation for Videos
- 5 Conclusion
- References
- Challenge to Scalability of Face Recognition Using Universal Eigenface
- 1 Introduction
- 2 Framework of Face Recognition by Weight Equations
- 2.1 Universal and Individual Eigenfaces
- 2.2 Face Recognition by Weight Equations
- 2.3 Scalability Problem with Face Recognition
- 3 Parallel Underdetermined Approach
- 3.1 Solution of Underdetermined Weight Equations
- 3.2 Parallel Underdetermined Systems
- 3.3 Preliminary Elimination by Parallel Underdetermined System
- 4 Experimental Challenge to Scalability
- 4.1 Construction of Scalable Database
- 4.2 Refinement of Universal Eigenface
- 4.3 Composition of Face Data
- 4.4 Specifications of Face Recognition Experiments
- 4.5 Fundamental Experiments of Face Recognition
- 4.6 Fundamental Experiments of Preliminary Elimination
- 4.7 Challenge to Scalability of Face Recognition
- 5 Conclusions
- References
- From Optimised Inpainting with Linear PDEs Towards Competitive Image Compression Codecs
- 1 Introduction
- 2 PDE-Based Inpainting
- 3 From Inpainting to Compression
- 4 Encoding with Exact Masks
- 5 Encoding with Stochastic Tree-Building
- 6 Experiments
- 7 Conclusion
- References
- A Study on Size Optimization of Scanned Textual Documents
- 1 Introduction
- 2 Contributing Factors to Large Scanned Document Size
- 3 Compression and Image Formats
- 3.1 Compression Techniques
- 3.2 Image Formats
- 4 Approach
- 5 Results and Discussions
- 6 Conclusion
- References
- Combination of Mean Shift of Colour Signature and Optical Flow for Tracking During Foreground and Background Occlusion
- 1 Introduction
- 2 Literature Study
- 3 The Proposed Model
- 3.1 Pre-localization
- 3.2 Filtering
- 3.3 Weighting
- 3.4 Updating
- 4 Evaluation Methodology
- 5 Result and Discussion
- 6 Conclusion
- References
- Rendered Benchmark Data Set for Evaluation of Occlusion-Handling Strategies of a Parts-Based Car Detector
- 1 Introduction
- 2 Rendered Benchmark Data Set
- 3 Parts-Based Car Detector
- 3.1 Extraction of Texture Descriptors
- 3.2 Learning of Parts-Based Object Representation
- 3.3 Parts-Based Detection Framework
- 4 Occlusion-Handling of the Parts-Based Car Detector
- 4.1 Contribution-Aware Strategy for Occlusion-Handling of a Parts-Based Car Detector
- 5 Conclusion and Further Work
- References
- Moving Object Detection Using Energy Model and Particle Filter for Dynamic Scene
- 1 Introduction
- 2 Overview
- 3 Initial Label Estimation
- 4 Tracking
- 5 Experiment
- 5.1 Experimental Environment
- 5.2 Performance Evaluation for Tracking
- 5.3 Performance of Challenging Environment
- 6 Conclusions
- References
- Logarithmically Improved Property Regression for Crowd Counting
- 1 Introduction
- 2 Literature Review
- 3 Proposed Approach
- 3.1 Textural Feature Selection
- 3.2 Geometric Feature Extraction for Segments
- 3.3 Perspective Correction
- 4 Gaussian Process Regression with Compounded Kernel
- 5 Experiments and Results
- 6 Conclusions
- References
- Lesioned-Part Identification by Classifying Entire-Body Gait Motions
- 1 Introduction
- 2 Related Work
- 3 Overview of the System
- 4 Gait Features for Lesioned-Part Identification
- 4.1 Dataset Collection for Lesioned-Part Identification
- 4.2 Gait Features Representing the Motion of the Entire Body
- 5 Experiments
- 6 Concluding Remarks
- References
- Variable-Length Segment Copy for Compressing Index Map of Palette Coding in Screen Content Coding
- Abstract
- 1 Introduction
- 2 Overview
- 3 Proposed Algoriths
- 3.1 Previous Line Matching Analysis and General 2-D Index-Segment Copy Scheme
- 3.2 1st Simplified Mode: Copy 2nd Above Mode
- 3.3 2nd Simplified Mode: Sliding Half Line Mode
- 3.4 3rd Simplified Mode: Splitting Half Line Mode
- 4 Experimental Results
- 5 Conclusions
- References
- Automatic Construction of Action Datasets Using Web Videos with Density-Based Cluster Analysis and Outlier Detection
- 1 Introduction
- 2 Related Work
- 3 Approach
- 3.1 Shot Collection
- 3.2 Shot Clustering
- 3.3 Shot Selection
- 4 Experiments and Results
- 4.1 Experimental Setup
- 4.2 Dataset Construction
- 4.3 Action Recognition
- 5 Conclusions
- References
- Image/Video Coding and Transmission
- Fast Coding Strategy for HEVC by Motion Features and Saliency Applied on Difference Between Successive Image Blocks
- 1 Introduction
- 2 Proposed Mode Selection Technique
- 2.1 Motion Features Extraction
- 2.2 Saliency Feature Extraction
- 2.3 Cost Function Based AOI Categorization
- 2.4 Intermode Selection
- 2.5 Threshold Determination
- 3 Experimental Results and Analysis
- 3.1 Experimental Setup
- 3.2 Results and Discussions
- 4 Conclusion
- References
- Neighboring Sample Prediction Coding for HEVC Screen Content Coding
- Abstract
- 1 Introduction
- 2 Palette Mode in HEVC SCC
- 3 Proposed Neighboring Sample Prediction Coding
- 4 Analysis for Neighboring Sample Prediction Coding
- 5 Signaling of Neighboring Sample Prediction Coding for HEVC SCC
- 6 Simulation Results
- 7 Conclusion
- References
- Computational Photography and Arts
- Aesthetic Interactive Hue Manipulation for Natural Scene Images
- 1 Introduction and Motivation
- 2 Related Work
- 2.1 Hue Manipulation
- 2.2 Image Editing Software
- 3 Hue Manipulation Framework
- 3.1 System Overview
- 3.2 Image Segmentation
- 3.3 Hue Histogram Selection
- 3.4 Hue Transfer Operators
- 4 Results
- 4.1 Hue Shift, Compress and Stretch
- 4.2 Combining the Three Operations
- 5 Discussion
- 5.1 Comparison with Other Software
- 5.2 User Feedback
- 6 Conclusion
- References
- Cross-View Action Recognition by Projection-Based Augmentation
- 1 Introduction
- 2 Related Work
- 3 Multi-projection-Based Framework
- 3.1 3D Action Decomposition
- 3.2 Feature Extraction
- 3.3 Decomposed Action Representation
- 4 Experiments
- 4.1 Prediction Procedure
- 4.2 Implementation Details
- 4.3 Northwestern-UCLA Multiview Action 3D Dataset
- 4.4 Action Recognition from the Seen Viewpoint
- 4.5 Action Recognition Across Different Viewpoints
- 5 Conclusion
- References
- Star-Effect Simulation for Photography Using Self-calibrated Stereo Vision
- 1 Introduction
- 2 Basic and Notation
- 2.1 Star Effects in Photography
- 2.2 Stereo Vision
- 3 Depth Estimation
- 4 Highlight Registration
- 4.1 Highlight Detection
- 4.2 Color Recovery
- 4.3 Luminance Estimation
- 5 Star Pattern Rendering
- 6 Experiments
- 7 Conclusion
- References
- Computer Vision and Applications
- A Robust Stereo Vision with Confidence Measure Based on Tree Agreement
- 1 Introduction
- 2 Related Work
- 3 Problem Statement
- 4 Confidence Measure with Tree Agreement
- 5 Proposed Method with Confidence Measure
- 5.1 Cost Aggregation Table
- 5.2 Cost Aggregation with Confidence Term
- 6 Experimental Results
- 6.1 Confidence Measure Comparison
- 6.2 Quantitative Evaluation on KITTI Dataset
- 6.3 Qualitative Evaluation on HCI Dataset
- 7 Conclusions
- References
- Semantics-Preserving Warping for Stereoscopic Image Retargeting
- 1 Introduction
- 2 Related Works
- 3 Our Approach
- 3.1 Pre-processing
- 3.2 Warping
- 3.3 Image Compositing
- 3.4 Optimization Details
- 4 Results and Discussion
- 5 Conclusion
- References
- Improved Poisson Surface Reconstruction with Various Passive Visual Cues from Multiple Camera Views
- 1 Introduction
- 2 Related Work
- 3 Proposed Mesh Optimization Method
- 3.1 One-Ring Neighborhood
- 3.2 Silhouette Consistency
- 3.3 Photometric Consistency
- 3.4 A Combinatorial Consistency
- 4 Experiments and Results
- 5 Conclusion
- References
- Prediction of Vibrations as a Measure of Terrain Traversability in Outdoor Structured and Natural Environments
- 1 Introduction
- 2 Problem Definition
- 2.1 Objective
- 2.2 Methodology
- 3 Regression of Motion Information
- 3.1 Terrain Patches Identification
- 3.2 Texture Attribute Extraction
- 3.3 Motion Feature Extraction
- 3.4 Matching Image and Motion Features
- 3.5 Regression Analysis using Gaussian Process (GP)
- 4 Experiment
- 4.1 Experiment Settings
- 4.2 Results
- 4.3 Discussion
- 5 Conclusion and Future Work
- References
- Echo State Network for 3D Motion Pattern Indexing: A Case Study on Tennis Forehands
- Abstract
- 1 Introduction
- 1.1 Related Work and Prior Studies
- 1.2 Tennis and Sport Science Backgrounds
- 2 Experimental Setup: Data Collection, Analysis, Pre-processing and ESN Modelling
- 2.1 Data Analysis and Pre-processing
- 2.2 Echo State Network -- Model Description and Parameter Optimisation Results
- 3 Classification Results and Model Visualisation
- 4 Discussion, Limitation and Critique
- 5 Conclusions, Recommendations and Future Work
- Acknowledgements
- References
- Image Segmentation and Classification
- Multispectral Image Denoising Using Optimized Vector NLM Filter
- 1 Introduction
- 2 Overview of NLM Filter
- 3 Optimized Vector NLM Filter
- 4 Experimental Results
- 5 Conclusion
- A Appendix
- References
- Scene-Based Non-uniformity Correction with Readout Noise Compensation
- 1 Introduction
- 2 Previous Work
- 3 Noise Model
- 3.1 Temporal Noise
- 3.2 Spatial Noise
- 4 Photometric Calibration
- 4.1 Camera Model
- 4.2 Non-uniformity Correction (NUC)
- 5 Readout Noise Compensation
- 6 Evaluation
- 6.1 Comparison to Ground Truth
- 6.2 Comparison to Evaluation Data Set
- 6.3 Longer Sequence
- 6.4 Results
- 7 Conclusion
- References
- A Color Quantization Based on Vector Error Diffusion and Particle Swarm Optimization Considering Human Visibility
- 1 Introduction
- 2 Conventional Color Quantization Methods
- 2.1 Median Cut Algorithm
- 2.2 K-means Clustering Algorithm
- 3 Proposed Method
- 3.1 Vector Error Diffusion
- 3.2 Generation of Color Palette by using Particle Swarm Optimization
- 4 Experimental Results
- 5 Conclusions
- References
- Fast Interactive Image Segmentation Using Bipartite Graph Based Random Walk with Restart
- 1 Introduction
- 2 Method
- 2.1 Graph Structure
- 2.2 Edge Weight Measurement
- 2.3 Naive RWR on Bipartite Graph
- 2.4 Accelerating RWR
- 3 Results
- 3.1 Qualitative Comparison
- 3.2 Quantitative Comparison and Error Estimation
- 3.3 Time Analysis
- 4 Conclusions
- References
- Adaptive Window Strategy for High-Speed and Robust KLT Feature Tracker
- 1 Introduction
- 2 Related Work
- 3 Overview of KLT Algorithm
- 3.1 Effect of Search Window Size for Rotation/Scaling
- 3.2 Implications of Fixed Search Window Size with Pyramidal KLT
- 3.3 Tracking Errors and KLT Iterations
- 4 Adaptive Window Size for KLT
- 5 Evaluations
- 6 Conclusions
- References
- Enhanced Phase Correlation for Reliable and Robust Estimation of Multiple Motion Distributions
- 1 Introduction
- 2 Approach
- 2.1 Structure Check
- 2.2 Spectral Significance Filtering
- 2.3 Delta Array Check
- 2.4 Delta Array Clustering
- 2.5 Multiresolution
- 3 Experiments
- 3.1 Middlebury Stereo Dataset
- 3.2 KITTI Optical Flow Dataset
- 4 Summary and Conclusion
- References
- Robust Visual Voice Activity Detection Using Long Short-Term Memory Recurrent Neural Network
- 1 Introduction
- 1.1 Related Works
- 2 Proposed Method
- 2.1 Face Detection
- 2.2 Face Tracking
- 2.3 Landmark Localization and Geometric Normalization
- 2.4 Centroid Distance Features
- 2.5 LSTM Recurrent Neural Network
- 2.6 Hidden Markov Model
- 3 Experiment Settings
- 3.1 Dataset
- 3.2 Network Architecture
- 3.3 HMM-based Classifier
- 4 Experimental Results
- 5 Conclusion
- References
- Wing-Surface Reconstruction of a Lanner-Falcon in Free Flapping Flight with Multiple Cameras
- 1 Introduction
- 2 Measurement Setup
- 3 Calibration
- 4 Epipolar Geometry
- 5 Calculation of the Displacement Field
- 6 Triangulation and Masking
- 7 Sequence of a Wing Flap
- 8 Upper and Lower Wing Side
- 9 Error Analysis
- 10 Conclusion
- References
- Underwater Active Oneshot Scan with Static Wave Pattern and Bundle Adjustment
- 1 Introduction
- 2 Related Work
- 3 Overview
- 3.1 System Configuration
- 3.2 Algorithm
- 3.3 Polynomial Approximation of Refraction
- 4 Depth Dependent Calibration
- 4.1 Overview of the Calibration Process
- 4.2 Sphere Based Projector Calibration
- 5 3D Reconstruction
- 5.1 Wave Grid Reconstruction
- 5.2 Refinement with Bundle Adjustment
- 6 Experiments
- 6.1 Depth Dependent Calibration
- 6.2 Wave Oneshot Reconstruction
- 6.3 Evaluation of Refinement Algorithm
- 7 Conclusion and Future Work
- References
- Using Image Features and Eye Tracking Device to Predict Human Emotions Towards Abstract Images
- 1 Introduction
- 2 Theories
- 2.1 Emotion
- 2.2 Abstract Art
- 2.3 Relevance Feedback
- 2.4 Support Vector Machine
- 3 Data Collection
- 4 Feature Extraction
- 4.1 Original Image
- 4.2 Original Image Processed with Eye Movement Data
- 4.3 Original Image Processed with Eye Movement Data and Gaussian Blur
- 5 Experiment Setup
- 6 Experimental Results and Discussion
- 6.1 User Model
- 6.2 Global Model - New User
- 7 Conclusion
- References
- Video Surveillance
- Personal Authentication Based on 3D Configuration of Micro-feature Points on Facial Surface
- 1 Introduction
- 2 Preliminary
- 2.1 Definition of Structural Similarity with Canonical Angles
- 2.2 Grassmann Discriminant Analysis
- 3 Proposed Framework
- 3.1 Feature Extraction with Separability Filter
- 3.2 Corresponding Feature Points using Autocorrelation Matrix
- 3.3 Overall Procedure of the Proposed Framework
- 4 Experimental Results and Consideration
- 4.1 Experiment 1
- 4.2 Experiment 2
- 4.3 Experiment 3
- 4.4 Experiment 4
- 5 Concluding Remarks
- References
- 6-DOF Direct Homography Tracking with Extended Kalman Filter
- 1 Introduction
- 2 ECC-Based 6-DOF Direct Homography Tracking
- 2.1 Homography from a 6-DOF Pose
- 2.2 ECC-based Direct Homography Tracking
- 3 Integration of Extended Kalman Filter
- 3.1 Translational Motion Model
- 3.2 Rotational Motion Model
- 4 Results
- 4.1 Pose Estimation
- 4.2 Speed Evaluation
- 4.3 Benchmark Experiments
- 5 Conclusion
- References
- Tracking a Human Fast and Reliably Against Occlusion and Human-Crossing
- 1 Introduction
- 2 Kernelized Correlation Filters
- 2.1 Circulant Matrices
- 2.2 Fast Kernel Regression
- 2.3 Fast Detection
- 2.4 Fast Kernel Correlation
- 3 Kalman Filter
- 4 The Proposed Method
- 5 Experiments
- 5.1 Implementation Details
- 5.2 Evaluation
- 6 Conclusions
- References
- Biomedical Image Processing and Analysis
- Automatic BI-RADS Classification of Mammograms
- 1 Introduction
- 2 Image Measures
- 2.1 Texture Measures from the Image
- 2.2 Texture Measures from Gray-Level Co-occurrence Matrix
- 2.3 VolparaTM Algorithm
- 3 Dataset and Quality Metrics
- 3.1 Quality-Score (QS)
- 3.2 Overlap-Area (OA)
- 4 Results
- 5 Conclusion
- References
- Analyzing Muscle Activity and Force with Skin Shape Captured by Non-contact Visual Sensor
- 1 Introduction
- 2 Related Work
- 3 Data Acquisition
- 4 Feature Extraction of Skin Deformation
- 4.1 Defining Feature Vector to Explain Skin Deformation
- 4.2 Finding Correspondence Between the Template Shape and Each Range Scan
- 5 Learning the Relationship Between Skin Shape, Force, and Muscle Activity
- 5.1 Estimating Force from Skin Shape
- 5.2 Estimating Muscle Activity from Skin Shape
- 5.3 Synthesizing Skin Shape from Muscle Activity
- 6 Experiments
- 7 Conclusion
- References
- Regression as a Tool to Measure Segmentation Quality and Preliminary Indicator of Diseased Lungs
- 1 Introduction
- 2 Data Collection
- 3 Methodology
- 3.1 Segmentation Algorithm
- 3.2 Regression Analysis
- 4 Results
- 5 Discussion
- 6 Conclusion
- References
- An Image Registration Method with Radial Feature Points Sampling: Application to Follow-Up CT Scans of a Solitary Pulmonary Nodule
- 1 Introduction
- 2 Outline of a Support System for Follow-Up CT Scans
- 3 Existing Method of the Feature Based Image Registration and Its Problem
- 3.1 Existing Method of Feature Based Image Registration
- 3.2 Problems of the Existing Method
- 4 Proposed Method
- 4.1 Selecting Feature Points
- 4.2 Sampling Matching Pairs
- 5 Experiments
- 5.1 Evaluation Data and Evaluation Method
- 5.2 Experimental Results
- 5.3 Discussion
- 6 Conclusion
- References
- Object and Pattern Recognition
- Time Consistent Estimation of End-Effectors from RGB-D Data
- 1 Introduction
- 2 Related Work
- 3 Input Data
- 4 Single Frame End-Effector Estimation
- 4.1 Point Cloud Topology Description
- 4.2 Estimating End-Effectors in a Topologically Weighted Graph
- 5 Temporal Coherence Guided End-Effector Estimation
- 5.1 Predict Phase
- 5.2 Update Phase
- 6 Experimental Results
- 6.1 Quantitative Results
- 6.2 Qualitative Result
- 7 Conclusion
- References
- Volume-Based Semantic Labeling with Signed Distance Functions
- 1 Introduction
- 2 Related Works
- 3 Description of the Method
- 3.1 Labeled TSDF
- 3.2 Volume Update Process
- 4 Experimental Evaluation
- 4.1 Robustness to Synthetic Label Noise
- 4.2 Results in Real Settings
- 5 Final Remarks
- References
- Simultaneous Camera, Light Position and Radiant Intensity Distribution Calibration
- 1 Introduction
- 1.1 Related Work
- 2 Shading of Lambertian Planes Under Near Illumination
- 2.1 Geometric Properties of Illumination Model
- 3 Illuminant Properties Estimation
- 3.1 Dominant Light Axis Estimation
- 3.2 Closed-Form Estimation of Light Position and RID
- 3.3 Optimisation Procedure for Complex Anisotropic Sources
- 4 Results
- 4.1 Synthetic Data
- 4.2 Real Data
- 5 Conclusions
- References
- A General Vocabulary Based Approach for Fine-Grained Object Recognition
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Features
- 3.2 Classifier
- 4 Experiment
- 4.1 Datasets
- 4.2 Results
- 5 Discussion and Future Work
- References
- A Triangle Mesh Reconstruction Method Taking into Account Silhouette Images
- 1 Introduction
- 2 Related Work
- 3 Overview of the Multi-view Projector Camera System
- 3.1 System Configuration
- 3.2 The Multi-view Projector Camera System
- 3.3 Creating Silhouette Based on the Database
- 4 Triangle Mesh Reconstruction
- 5 Experiment
- 5.1 Merging Point Sets
- 5.2 Computing the Delaunay Triangulation in 3D Space
- 5.3 Graph Cut Optimization
- 5.4 Experimental Settings
- 5.5 Experimental Results
- 6 Conclusion
- References
- All-Focus Image Fusion and Depth Image Estimation Based on Iterative Splitting Technique for Multi-focus Images
- Abstract
- 1 Introduction
- 2 Iterative Region-Splitting Algorithm
- 2.1 Initial All-Focus Image Generation
- 2.2 Initial Region Segmentation
- 2.3 Iterative Region Splitting
- 2.3.1 Classification of a Region S
- 2.3.1 Classification of a Region S
- 2.3.2 Region Splitting
- 2.3.3 Processing of X-Regions
- 3 Depth Image Post-processing
- 4 Experiment Results
- 5 Conclusion
- References
- Stereo Matching Techniques for High Dynamic Range Image Pairs
- 1 Introduction
- 2 HDR Images and Stereo Matching
- 3 Algorithm and Evaluation
- 4 Conclusion
- References
- Discriminative Properties in Directional Distributions for Image Pattern Recognition
- 1 Introduction
- 2 Mathematical Preliminaries
- 2.1 Directions and Matching
- 2.2 Aggregating Methods for Local Regions
- 2.3 Histogram of Gradients
- 2.4 Histogram of Dominant Directions
- 2.5 Gradient-Based Object Detection Method
- 2.6 Wasserstein Distance
- 3 Feature Extractions and Measurements
- 3.1 Local Directional Distributions Methods
- 3.2 Global Directional Distribution Method
- 3.3 Histogram of Oriented Gradients Method
- 4 Numerical Experiments
- 5 Discussion
- 6 Conclusions
- References
- Deep Boltzmann Machines for i-Vector Based Audio-Visual Person Identification
- 1 Introduction
- 2 Background
- 2.1 Deep Boltzmann Machines
- 2.2 Total Variability Modeling (TVM)
- 3 DBM-DNN Classification
- 4 Database and Features
- 4.1 Speech Features
- 4.2 Visual Features
- 5 Results and Analysis
- 5.1 Implementation
- 6 Conclusion
- References
- Improved DSIFT Descriptor Based Copy-Rotate-Move Forgery Detection
- 1 Introduction
- 2 Related Works
- 3 Forgery Detection Algorithms
- 3.1 Steps to Improve DSIFT
- 3.2 CMF Detection with Translation/Rotation
- 3.3 False Match Removal
- 3.4 Neighbourhood Clustering
- 4 Experiments and Evaluation Method
- 4.1 Datasets and Evaluation Method
- 4.2 Experiments and Results for CMF/CRM Forgery Detection
- 4.3 An Experiment to Test Rotation Invariant of Improved DSIFT
- 4.4 An Experiment to Find the Difference Between Matching Points Algorithms in CMF Detection
- 5 Conclusion
- References
- Local Clustering Patterns in Polar Coordinate for Face Recognition
- 1 Introduction
- 2 The Proposed Local Pattern Descriptor
- 2.1 Local Clustering Pattern (LCP)
- 2.2 Coding Scheme
- 3 Experimental Results
- 3.1 Evaluation of Similarity
- 3.2 Experimental Results of LCP in Various Coordinate System
- 3.3 Experimental Results on Extended Yale B Database
- 3.4 Experimental Results on CAS-PEAL Database
- 4 Conclusion
- References
- Computer Vision and Pattern Recognition
- Deep Convolutional Neural Network in Deformable Part Models for Face Detection
- 1 Introduction
- 2 Related Works
- 3 Deep Face Deformable Part Models
- 3.1 New Face Representation Model
- 3.2 DeepFace DPM - A Convolutional Neural Network Integrated in DPM
- 4 Intuitive Non-maximum Suppression
- 5 Experimental Results
- 6 Conclusion
- References
- Multimodal Gesture Recognition Using Multi-stream Recurrent Neural Network
- 1 Introduction
- 2 Related Work
- 3 Multi-stream Recurrent Neural Network
- 3.1 RNN with LSTM Cells
- 3.2 Recurrent Multimodal Fusion
- 3.3 Late Multimodal Fusion
- 3.4 Early Multimodal Fusion
- 4 Experiments
- 4.1 Dataset
- 4.2 Network Architecture
- 4.3 Training Settings
- 4.4 Experimental Results
- 5 Conclusion
- References
- Image/Video Processing and Analysis
- A Spatially Constrained Asymmetric Gaussian Mixture Model for Image Segmentation
- 1 Introduction
- 2 Background
- 3 Proposed Algorithm
- 4 Experimental Results
- 5 Conclusion
- References
- Object Recognition in Baggage Inspection Using Adaptive Sparse Representations of X-ray Images
- 1 Introduction
- 2 Proposed Method
- 2.1 Model Learning
- 2.2 Testing
- 3 Experiments
- 4 Conclusions
- References
- Real-Time Lane Estimation Using Deep Features and Extra Trees Regression
- 1 Introduction
- 2 Literature Review
- 3 Algorithm
- 3.1 Training
- 3.2 Testing
- 4 Experimental Results
- 4.1 Comparative Analysis
- 4.2 Parameter Analysis
- 5 Conclusion and Future Work
- References
- Contrast Based Hierarchical Spatial-Temporal Saliency for Video
- 1 Introduction
- 2 Related Work
- 3 Hierarchical Spatial-Temporal Saliency Model
- 3.1 Saliency Entity Construction
- 3.2 Exploiting Motion Information
- 3.3 Spatial-Temporal Saliency Generation
- 4 Experimental Setup
- 4.1 Dataset
- 4.2 Evaluation Metrics
- 5 Evaluation of the Proposed Method
- 5.1 Evaluation of Introduction to Regional Characteristics
- 5.2 Evaluation of Adaptive Temporal Windows
- 6 Comparison with State-of-the-Art
- 6.1 Salient Object Detection
- 6.2 Eye Fixation Prediction
- 7 Conclusion
- References
- Pattern Recognition
- Binary Descriptor Based on Heat Diffusion for Non-rigid Shape Analysis
- 1 Introduction
- 2 Background
- 3 Proposed Method
- 3.1 Scalar Field Definition
- 3.2 Intrinsic Local Reference Frame
- 4 Local Binary Descriptor
- 5 Experiments
- 5.1 Dataset
- 5.2 Performance of the Local Polar Coordinate System
- 5.3 Performance of the Binary Descriptor
- 5.4 Parameter Selection
- 6 Conclusion
- References
- Table Detection from Slide Images
- 1 Introduction
- 2 Related Works
- 3 Detection of Rows and Columns
- 4 Table Area Positioning
- 4.1 Table Candidate Generation
- 4.2 Table Candidate Evaluation
- 4.3 Table Area Expansion
- 5 Table Confirmation
- 6 Evaluation
- 6.1 Datasets and Metrics
- 6.2 Experiments on Training and Benchmark Datasets
- 6.3 Evaluation on Test Dataset
- 7 Conclusion
- References
- Face Search in Encrypted Domain
- 1 Introduction
- 2 Related Work
- 3 Our Contributions
- 3.1 Image Encryptions
- 3.2 Face Object Search
- 3.3 Search Evaluations
- 4 Results and Analysis
- 5 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.