Pattern Recognition and Computer Vision

Name: Pattern Recognition and Computer Vision | Second Chinese Conference, PRCV 2019, Xi'an, China, November 8-11, 2019, Proceedings, Part III
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

Second Chinese Conference, PRCV 2019, Xi'an, China, November 8-11, 2019, Proceedings, Part III

Zhouchen Lin Liang Wang Jian Yang Guangming Shi Tieniu Tan Nanning Zheng Xilin Chen Yanning Zhang(Editor)

Springer (Publisher)

Published on 31. October 2019

XVIII, 545 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-31726-3 (ISBN)

€53.49incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
Contents - Part III
Data Analysis and Optimization
Modality Consistent Generative Adversarial Network for Cross-Modal Retrieval
Abstract
1 Introduction
1.1 Motivation and Contribution
2 Related Work
2.1 Non-GANs-Based Cross-Modal Retrieval Methods
2.2 Generative Adversarial Networks (GANs)-Based Cross-Modal Retrieval Methods
3 Our Approach
3.1 Problem Formulation
3.2 Generative Model
3.3 Discrimination Model
3.4 Modality Consistent Embedding Network
3.5 Optimization
4 Experiments
4.1 Datasets
4.2 Evaluation Measure and Compared Methods
4.3 Implementation Detail
4.4 Result and Discussion
5 Conclusion
Acknowledgements
References
Retrieval by Classification: Discriminative Binary Embedding for Sketch-Based Image Retrieval
1 Introduction
2 Related Work
3 Discriminative Binary Embedding
3.1 Notation and Problem Definition
3.2 Framework
4 Experiments
4.1 Implementation Details
4.2 Datasets
4.3 Evaluation Protocol and Baselines
4.4 Results and Discussions
5 Conclusion
References
Robust Subspace Segmentation via Sparse Relation Representation
1 Introduction
2 Preliminary
3 Motivation
3.1 Relation Representation
3.2 Sparse Relation Representation (SRR)
4 Optimization and Analyses
4.1 Optimization
4.2 Algorithm
4.3 Analyses
5 Experiments
5.1 Experiments on Hopkins 155 Database
5.2 Experiments on Face Image Databases
6 Conclusion
References
An Approach to the Applicability Evaluation of Moving Target Tracking Algorithm
Abstract
1 Introduction
2 Datasets
2.1 Test Factors and Levels
2.2 Selection of Image Sequences
2.3 Organization of Image Data Selection
2.4 Test Dataset Structure
3 Orthogonal Test Evaluation Method
3.1 Evaluation Framework
3.2 Selection and Construction of Evaluation Indicators
4 Experimental Results and Analysis
4.1 Range Analysis
4.2 Analysis of Test Results
5 Conclusion and Future Directions
Acknowledgement
References
A Cooperative Particle Swarm Optimization Algorithm Based on Greedy Disturbance
Abstract
1 Introduction
2 GCPSO Algorithm
3 Experiments and Analysis
3.1 Quality Analysis of Solutions
3.2 Convergence Analysis of the Algorithm
4 Conclusion
References
Jointing Cross-Modality Retrieval to Reweight Attributes for Image Caption Generation
1 Introduction
2 Our Model
2.1 Image-Caption Retrieval
2.2 The General and Specific Semantics Jointed Visual Feature Extraction
2.3 Image Caption Generator
3 Experiments and Results
3.1 Datasets and Experimental Setup
3.2 Results and Analysis
4 Conclusions
References
Pseudo Label Guided Subspace Learning for Multi-view Data
1 Introduction
2 Related Work
2.1 Spectral Clustering
2.2 Clustering with Adaptive Neighbors
3 Methodology
3.1 Model
3.2 Optimization
4 Experiments
4.1 Datasets and Settings
4.2 Experimental Results
4.3 Parameter Sensitivity
5 Conclusion
References
MVB: A Large-Scale Dataset for Baggage Re-Identification and Merged Siamese Networks
1 Introduction
2 Dataset
2.1 Raw Data Collection System
2.2 Data Annotation
2.3 Dataset Characteristics
3 Task and Evaluation Metric
4 Baseline Method
4.1 Basic Siamese Network
4.2 Merged Siamese Network
5 Experiments
5.1 Performance and Ablation Study
5.2 Case Study
5.3 Future Work
6 Conclusion
References
Personalized Travel Recommendation via Multi-view Representation Learning
Abstract
1 Introduction
2 Related Work
2.1 Deep Learning Based Recommender System
2.2 Personalized Travel Recommendation
3 The Proposed PTRMRL Framework
3.1 Problem Definition
3.2 Architecture
3.3 Multi-view User Preferences Representation
3.4 The Attention Network
4 Experiments
4.1 Dataset and Evaluation Protocols
4.2 The Impact of Embedding Dimension
4.3 Overall Comparison
4.4 The Impact of Single-View Data on Recommendation Effects
5 Conclusion
References
FollowMeUp Sports: New Benchmark for 2D Human Keypoint Recognition
1 Introduction
2 Related Work
3 The Dataset
3.1 Pose Estimation
4 Analysis of the State of the Art
4.1 Comparisons of Approaches on the COCO Dataset
4.2 Comparisons of Approaches on the FollowMeUp Dataset
4.3 The Effect of Training on the FollowMeUp Dataset
5 Conclusion
References
Partial Order Structure Based Image Retrieval
Abstract
1 Introduction
2 Related Works
3 Image Retrieval Based on Partial Order Structure
3.1 Visual Semantics
3.2 Object Partial Order Structure
3.3 Attribute Partial Order Structure
4 Experiments
4.1 Dataset
4.2 Setup
4.3 Experimental Results
5 Conclusion
Acknowledgements
References
Computer Vision Applications
Semantic Object and Plane SLAM for RGB-D Cameras
1 Introduction
2 Related Work
2.1 Object Detection
2.2 Object SLAM
2.3 Planar SLAM
3 Method
3.1 System Overview
3.2 Object Mapping
3.3 Plane Mapping
3.4 Bundle Adjustment with Points, Objects and Planes
4 Experiment
4.1 Experimental Settings
4.2 Results
5 Conclusions
References
Crime Scene Sketches Classification Based on CNN
1 Introduction
2 Related Works
3 Crime Scene Sketch
3.1 Sketch Taxonomy
3.2 Rules of Uploading Sketch
4 Methodology
5 Experiment
5.1 Data Set
5.2 Architecture
6 Results and Analysis
6.1 Compared with Classic CNNs
6.2 Performance Analysis
7 Conclusion
References
Image-Based Air Quality Estimation
1 Introduction
2 Related Work
3 Our Database
3.1 Collection and Annotation
3.2 Appearance
3.3 Diversity
4 Image Based Air Quality Estimation
5 Conclusion
References
Rotational Alignment of IMU-camera Systems with 1-Point RANSAC
1 Introduction
2 Homography Constraints
3 IMU-camera Calibration
3.1 1pt-RANSAC Calibration Method
3.2 Non-linear Parameter Optimization
4 Experiments
4.1 Experiments with Synthetic Data
4.2 Real Scene Data Experiment
5 Conclusion
References
Bidirectional Adversarial Domain Adaptation with Semantic Consistency
1 Introduction
2 Related Work
3 The Proposed Model
3.1 Formulations
3.2 Optimization
4 Experiments
4.1 Datasets
4.2 Experimental Setup
4.3 Comparison with Existing Methods
4.4 Evaluation on Semantic Consistency
4.5 Ablation Study
5 Conclusion
References
A Novel Hard Mining Center-Triplet Loss for Person Re-identification
Abstract
1 Introduction
2 Related Work
3 Proposed Hard Mining Center-Triplet Loss
3.1 Hard Mining Center-Triplet Loss Function
3.2 Our Deep Convolutional Neural Network Architecture
3.3 Compared with Other Loss Functions
4 Experiments
4.1 Datasets
4.2 Implementation Details
4.3 Experimental Results and Analysis
5 Conclusion
Acknowledgement
References
Kinematic Feature-Based Evaluation Method for Elderly Balance Ability by Using Factor Analysis
Abstract
1 Introduction
2 Data Description and Extraction of Feature Indicators
2.1 Experimental Data
2.2 Feature Indicators Extraction Based on the Kinematics
3 Kinematic Feature-Based Evaluation Method
3.1 Elderly Balance Ability Factor Selection and Calculation Method
3.2 Comprehensive Evaluation Scoring Model for Elderly Balance Ability
4 Results and Discussion
4.1 Analysis of Elderly Balance Ability Factors and Score Results
4.2 Elderly Balance Ability Comprehensive Score Results
5 Conclusion
References
Efficient Automatic Meta Optimization Search for Few-Shot Learning
1 Introduction
2 Related Work
2.1 Meta Learning
2.2 Neural Architecture Search
3 Method
3.1 Generating Transferable Architecture by Controller
3.2 Training Controller with Reinforcement Learning
3.3 Training Child Model with Reptile
4 Experiment
4.1 Few-Shot Learning Datasets
4.2 Implementation Details
4.3 Evaluation
5 Conclusion
References
Visual Odometry with Deep Bidirectional Recurrent Neural Networks
1 Introduction
2 Related Work
3 Approach
3.1 Encoder
3.2 Forward-Backward Process with 2D Convolutional LSTM
3.3 Coupled Local and Global Constraints
4 Experiments
4.1 Implementation
4.2 Dataset
4.3 Ablation Study
4.4 Results on KITTI Dataset
4.5 Results on Málaga Dataset
5 Conclusion
References
Fuzzy Control Reversing System Based on Visual Information
1 Introduction
2 Related Work
3 Method
3.1 Vision-Based Reversing Model
3.2 Real-Time Identification of Obstacles
3.3 Detection and Exclusion of Dynamic Obstacle Feature Points
3.4 Obstacle Avoidance and Path Planning
4 Experiments
4.1 Reverse Simulation Experiment
4.2 Detection Results of Feature Points in the Image
5 Conclusion
References
Adversarial Domain Alignment Feature Similarity Enhancement Learning for Unsupervised Domain Adaptation
1 Introduction
1.1 Motivation and Contribution
2 Related Work
3 Proposed Approach
3.1 Domain Adversarial Training
3.2 Feature Extraction and Label Classifier Learning
3.3 Optimization with Backpropagation
4 Experiments
4.1 Datasets
4.2 Experimental Settings
4.3 Results and Analysis
4.4 Discussion
5 Conclusion
References
ADSRNet: Attention-Based Densely Connected Network for Image Super-Resolution
1 Introduction
2 Related Work
3 Proposed Method
3.1 Network Architecture
3.2 Dense Attention Module
3.3 Channel Attention Module
3.4 Mapping Function
4 Experiments
4.1 Datasets and Metrics
4.2 Implementation Details
4.3 Model Analysis
4.4 Comparisons with the State-of-the-Arts
5 Conclusion
References
Robust and Efficient Visual-Inertial Odometry with Multi-plane Priors
1 Introduction
2 Related Works
3 Visual-Inertial Odometry with Multi-plane Priors
3.1 Initialization and Plane Detection
3.2 Visual-Inertial-Plane PnP Tracking
3.3 Plane Expansion via Reprojection Consensus
3.4 Sliding-Window Optimization with Structureless Plane-Distance Cost
4 Experiments
4.1 Tracking Accuracy and Robustness
4.2 Efficiency
5 Conclusions and Disscusions
References
Contour-Guided Person Re-identification
1 Introduction
2 Related Work
3 Contour-Guided Architecture
3.1 Plain Architecture
3.2 Multi-scale Architecture
3.3 Contour-Guided Auxiliary Mechanism
4 Experiment
4.1 Datasets and Evaluation Protocol
4.2 Implementation Details
4.3 Performance Evaluation
5 Conclusion
References
Robust License Plate Detection Through Auxiliary Information and Context Fusion Model
1 Introduction
2 Related Work
2.1 Deep Learning for License Plate Detection
2.2 Available Information for License Plate Detection
3 Proposed Method
3.1 Auxiliary Information Enhancement
3.2 Context Fusion Model
3.3 Loss Function
4 Experiments
4.1 License Plate Datasets
4.2 Implementation Details
4.3 Experimental Results
5 Conclusion
References
PointNet-Based Channel Attention VLAD Network
1 Introduction
2 Related Work
3 Method
3.1 Channel Attention Module
3.2 NetVLAD Module
3.3 Proof of Symmetry of NetVLAD
4 Experiments
4.1 3D Object Classification
4.2 3D Semantic Segmentation
5 Conclusion
References
Multi-scale Deep Residual Network for Satellite Image Super-Resolution Reconstruction
Abstract
1 Introduction
2 Related Work
3 Proposed Method
3.1 Symmetric Convolution and Deconvolution
3.2 Multi-scale Module
3.3 Architecture
4 Experiments
4.1 Datasets
4.2 Implementation Details
4.3 Results
5 Conclusion
References
CG Animation Creator: Auto-rendering of Motion Stick Figure Based on Conditional Adversarial Learning
1 Introduction
2 Related Work
3 Animation Rendering System Based on High-Resolution Conditional Learning Network with Semantic Information
3.1 Database Creation for System Training
3.2 Pix2pixHD for High-Resolution Image Synthesis and Semantic Manipulation
4 Experimental Results and Analysis
4.1 The Experimental Setups
4.2 The Similarity Evaluation of Rendered Animation Image
4.3 The Rendering Loss and Nash Equilibrium of the Model
4.4 The Time Efficiency of the Automatic Rendering System
5 Conclusion
References
Deep Eyes: Binocular Depth-from-Focus on Focal Stack Pairs
1 Introduction
2 Related Work
3 Dual Focal Stack Dataset
4 B-DfF Network Architecture
4.1 Focus-Net and Focus-Net-Guided for DfF
4.2 Stereo-Net and BDfF-Net for Depth from Binocular Focal Stack
5 Implementation
6 Experiments
6.1 Extract the EDoF Image from Focal Stack
6.2 Depth Estimation from Focal Stack
6.3 Depth Estimation from Stereo and Binocular Focal Stack
6.4 Real Scene Experiment
7 Discussions and Future Work
References
Small Defect Detection in Industrial X-Ray Using Convolutional Neural Network
Abstract
1 Introduction
1.1 Small Defect Data Augmentation
1.2 Low-Level Features Detection
1.3 Multi-scale Transform Module
2 Related Work
2.1 RetinaNet
2.2 FPN
3 Data
3.1 Datasets
3.2 Metrics
4 Methods
4.1 Data Augmentation
4.2 Low-Level Features Extraction
4.3 Multi-scale Transform Module
5 Experiments
5.1 Implementation Details
5.2 Ablation Studies
5.3 Comparison with Other Models
5.4 Comparison with X-Ray Machine
6 Conclusion
References
ODCN: Optimized Dilated Convolution Network for 3D Shape Segmentation
Abstract
1 Introduction
2 Related Work
3 Proposed Method
3.1 Input
3.2 ODCN Structure
4 Experimental Setup and Evaluation
4.1 Datasets and Implementation Details
4.2 Comparison Between Different Methods
4.3 Analysis of the ODCN
5 Conclusion
Acknowledgments
References
Style Consistency Constrained Fusion Feature Learning for Liver Tumor Segmentation
Abstract
1 Introduction
2 Related Work
3 Proposed Method
3.1 Style Consistency Loss
3.2 Segmentation Network
4 Experiments
4.1 Datasets
4.2 Experiment Settings
4.3 Comparison Methods
4.4 Evaluation Metrics
4.5 Results and Analysis
5 Conclusion
Acknowledgement
References
Hierarchical Correlation Stereo Matching Network
1 Introduction
2 Related Work
3 Approach
3.1 Network Architecture
3.2 Stacked Hourglass Feature Network
3.3 Cost Volume
3.4 Cost Aggregation Network
3.5 Loss Function
4 Experiments
4.1 Datasets and Metrics
4.2 Implementation Details
4.3 Ablation Study
4.4 KITTI Test Results
5 Conclusion
References
An Accurate LSTM Based Video Heart Rate Estimation Method
1 Introduction
2 Related Work
3 Proposed Approach
3.1 Signal Preprocessing
3.2 LSTM Filtering Model
3.3 Training Strategy
4 Experiments
4.1 Experimental Settings
4.2 Experimental Results
5 Conclusion and Further Work
References
Self-supervised Homography Prediction CNN for Accurate Lane Marking Fitting
1 Introduction
2 Methods
2.1 Deep Network for Homography Predition
2.2 Loss Function and Self-supervised Learning
3 Experiments
3.1 Dataset
3.2 Training Setup and Testing
3.3 Evaluation and Results
4 Conclusion
References
Scenario Referring Expression Comprehension via Attributes of Vision and Language
1 Introduction
2 Related Work
2.1 Image and Text Retrieval
2.2 Referring Expression Comprehension
2.3 Scenario Image Dataset
3 Approach
3.1 Scenario Dataset Enhancement
3.2 Expression Attributes Extraction
3.3 Image Attributes Extraction
3.4 Attributes Retrieval Matching
4 Experiments
4.1 Single-Sequence Comprehension
4.2 Multi-sequences Comprehension
4.3 Results and Analysis
5 Conclusion
References
Incremental Poisson Surface Reconstruction for Large Scale Three-Dimensional Modeling
1 Introduction
2 Related Work
2.1 Surface Reconstruction
2.2 Poisson Surface Reconstruction
2.3 Incremental Surface Reconstruction
3 The Proposed Method
3.1 Motivation
3.2 Point Cloud Partition
3.3 Octree Nodes Classification
3.4 Incremental Reconstruction with Boundary Constraints
4 Experiments
4.1 Reconstruction on Benchmark Dataset
4.2 Reconstruction on Landscape Dataset
5 Conclusion
References
Deep Voice-Visual Cross-Modal Retrieval with Deep Feature Similarity Learning
1 Introduction
2 Prior Work
3 The Proposed Method
3.1 Problem Definition
3.2 Multimodal Architecture
3.3 Joining the Voice and Image Branches
4 Experiments
4.1 Dataset and Evaluation Protocols
4.2 Implementation Detail
4.3 Evaluation of Different Factors
4.4 Method Comparison
5 Conclusion
References
Exploiting Human Pose for Weakly-Supervised Temporal Action Localization
1 Introduction
2 Related Work
3 Method
3.1 Feature Extraction
3.2 Pose Pattern Sequence Module
3.3 Class Activation Sequence Module
3.4 Post Processing
4 Experiment
4.1 Dataset
4.2 Implementation Details
4.3 Ablation Study
4.4 Comparison with the State-of-the-Art Methods
5 Conclusion
References
Combing Deep and Handcrafted Features for NTV-NRPCA Based Fabric Defect Detection
1 Introduction
2 Proposed Method
2.1 Deep-Handcrafted Feature Extraction
2.2 Model Construction
2.3 The Generation and Segmentation of the Saliency Map
3 Experiments
3.1 Implementation Details
3.2 Comparison with State-of-the-Art
4 Conclusion
References
A Cost-Sensitive Shared Hidden Layer Autoencoder for Cross-Project Defect Prediction
Abstract
1 Introduction
2 Related Work
2.1 Cross-Project Defect Prediction
2.2 Deep Learning
3 Proposed Methodology
3.1 Data Normalization
3.2 Feature Extraction Model
3.3 Cost-Sensitive Softmax Classifier Model
4 Experiment
4.1 Datasets
4.2 Evaluation Metrics
4.3 Implementation Detail
4.4 Experiment Setup
4.5 Experiment Result and Analysis
5 Conclusion
Acknowledgements
References
Person ReID: Optimization of Domain Adaption Though Clothing Style Transfer Between Datasets
Abstract
1 Introduction
2 Related Works
2.1 Semantic Segmentation of Pedestrian Mask
2.2 Image-to-Image Translation with GAN
2.3 GAN in Person ReID Research
3 Implementation
3.1 Clothing Mask Extraction
3.2 Clothing Style Transfer
4 Experiment
4.1 Implement Details
4.2 Clothing Mask Extraction
4.3 Clothing Style Transfer
4.4 Tests of Domain Adaption on Transferred Datasets
4.5 Ablation Study
5 Conclusions
Acknowledgement
References
Shellfish Detection Based on Fusion Attention Mechanism in End-to-End Network
Abstract
1 Introduction
2 Methodology
3 Experiments
3.1 Dataset
3.2 Data Augmentation
3.3 Training Procedure
4 Results and Analysis
5 Conclusions and Future Works
Acknowledgements
References
Multi-branch Structure for Hierarchical Classification in Plant Disease Recognition
1 Introduction
2 Related Work
3 Multi-branch Hierarchical Classification Model
3.1 The Inter-relation of Diseases Across Species
3.2 Model Architecture
3.3 Generalized Model
4 Experimental Results
5 Conclusion
References
Correction to: Personalized Travel Recommendation via Multi-view Representation Learning
Correction to: Chapter "Personalized Travel Recommendation via Multi-view Representation Learning" in: Z. Lin et al. (Eds.): Pattern Recognition and Computer Vision, LNCS 11859, https://doi.org/10.1007/978-3-030-31726-3_9
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Pattern Recognition and Computer Vision

Description

More details

Other editions

Additional editions

Content

System requirements