Pattern Recognition and Computer Vision : 6th Chinese Conference, PRCV 2023, Xiamen, China, October 13–15, 2023, Proceedings, Part X / edited by Qingshan Liu, Hanzi Wang, Zhanyu Ma, Weishi Zheng, Hongbin Zha, Xilin Chen, Liang Wang, Rongrong Ji.

Format:

Book

Author/Creator:

Liu, Qingshan.

Contributor:

Wang, Hanzi.

Ma, Zhanyu.

Zheng, Weishi.

Zha, Hongbin.

Chen, Xilin.

Wang, Liang.

Ji, Rongrong.

Series:

Lecture Notes in Computer Science, 1611-3349 ; 14434

Language:

English

Subjects (All):

Image processing--Digital techniques.

Image processing.

Computer vision.

Artificial intelligence.

Application software.

Computer networks.

Computer systems.

Machine learning.

Computer Imaging, Vision, Pattern Recognition and Graphics.

Artificial Intelligence.

Computer and Information Systems Applications.

Computer Communication Networks.

Computer System Implementation.

Machine Learning.

Local Subjects:

Computer Imaging, Vision, Pattern Recognition and Graphics.

Artificial Intelligence.

Computer and Information Systems Applications.

Computer Communication Networks.

Computer System Implementation.

Machine Learning.

Physical Description:

1 online resource (509 pages)

Edition:

1st ed. 2024.

Place of Publication:

Singapore : Springer Nature Singapore : Imprint: Springer, 2024.

Summary:

The 13-volume set LNCS 14425-14437 constitutes the refereed proceedings of the 6th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2023, held in Xiamen, China, during October 13–15, 2023. The 532 full papers presented in these volumes were selected from 1420 submissions. The papers have been organized in the following topical sections: Action Recognition, Multi-Modal Information Processing, 3D Vision and Reconstruction, Character Recognition, Fundamental Theory of Computer Vision, Machine Learning, Vision Problems in Robotics, Autonomous Driving, Pattern Classification and Cluster Analysis, Performance Evaluation and Benchmarks, Remote Sensing Image Interpretation, Biometric Recognition, Face Recognition and Pose Recognition, Structural Pattern Recognition, Computational Photography, Sensing and Display Technology, Video Analysis and Understanding, Vision Applications and Systems, Document Analysis and Recognition, Feature Extraction and Feature Selection, Multimedia Analysis and Reasoning, Optimization and Learning methods, Neural Network and Deep Learning, Low-Level Vision and Image Processing, Object Detection, Tracking and Identification, Medical Image Processing and Analysis. .

Contents:

Intro

Preface

Organization

Contents - Part X

Neural Network and Deep Learning III

Dual-Stream Context-Aware Neural Network for Survival Prediction from Whole Slide Images

1 Introduction

2 Method

3 Experiments and Results

4 Conclusion

References

A Multi-label Image Recognition Algorithm Based on Spatial and Semantic Correlation Interaction

2 Related Work

2.1 Correlation-Agnostic Algorithms

2.2 Spatial Correlation Algorithms

2.3 Semantic Correlation Algorithms

3 Methodology

3.1 Definition of Multi-label Image Recognition

3.2 The Framework of SSCI

3.3 Loss Function

4 Experiments

4.1 Evaluation Metrics

4.2 Implementation Details

4.3 Comparison with Other Mainstream Algorithms

4.4 Evaluation of the SSCI Effectiveness

5 Conclusion

Hierarchical Spatial-Temporal Network for Skeleton-Based Temporal Action Segmentation

2.1 Temporal Action Segmentation

2.2 Skeleton-Based Action Recognition

3 Method

3.1 Network Architecture

3.2 Multi-Branch Transfer Fusion Module

3.3 Multi-Scale Temporal Convolution Module

3.4 Loss Function

4.1 Setup

4.2 Effect of Hierarchical Model

4.3 Effect of Multiple Modalties

4.4 Effect of Multi-modal Fusion Methods

4.5 Effect of Multi-Scale Temporal Convolution

4.6 Comparision with State-of-the-Art

Multi-behavior Enhanced Graph Neural Networks for Social Recommendation

3 Preliminaries

4 Methodology

4.1 Embedding Layer

4.2 Propagation Layer

4.3 Multi-behavior Integration Layer

4.4 Prediction Layer

4.5 Model Training

5 Experiments

5.1 Experimental Settings

5.2 Performance Comparison (RQ1)

5.3 Ablation Study (RQ2).

5.4 Parameter Analysis (RQ3)

6 Conclusion and Future Work

A Complex-Valued Neural Network Based Robust Image Compression

2 Related Works

2.1 Neural Image Compression

2.2 Adversarial Attack

2.3 Complex-Valued Convolutional Neural Networks

3 Proposed Method

3.1 Overall Framework

3.2 Nonlinear Transform

4 Experiment Results

4.1 Experiment Setup

4.2 Results and Comparison

4.3 Ablation Study

5 Conclusions

Binarizing Super-Resolution Neural Network Without Batch Normalization

3.1 Batch Normalization in SR Models

3.2 Channel-Wise Asymmetric Binarizer for Activations

3.3 Smoothness-Controlled Estimator

4 Experimentation

4.2 Ablation Study

4.3 Visualization

Infrared and Visible Image Fusion via Test-Time Training

2.1 Overall Framework

2.2 Training and Testing

3 Experiments

3.1 Experiment Configuration

3.2 Performance Comparison on TNO

3.3 Performance Comparison on VIFB

3.4 Ablation Study

Graph-Based Dependency-Aware Non-Intrusive Load Monitoring

2 Proposed Method

2.1 Problem Formulation

2.2 Co-occurrence Probability Graph

2.3 Graph Structure Learning

2.4 Graph Attention Neural Network

2.5 Encoder-Decoder Module

3 Numerical Studies and Discussions

3.1 Dataset and Experiment Setup

3.2 Metrics and Comparisons

Few-Shot Object Detection via Classify-Free RPN

2.1 Object Detection

2.2 Few-Shot Learning

2.3 Few-Shot Object Detection

3.1 Problem Setting

3.2 Analysis of the Base Class Bias Issue in RPN

3.3 Classify-Free RPN.

4 Experiments

4.1 Experimental Setup

4.2 Comparison with the State-of-the-Art

IPFR: Identity-Preserving Face Reenactment with Enhanced Domain Adversarial Training and Multi-level Identity Priors

2 Methods

2.1 Target Motion Encoder and 3D Shape Encoder

2.2 3D Shape-Aware Warping Module

2.3 Identity-Aware Refining Module

2.4 Enhanced Domain Discriminator

2.5 Training

3 Experiment

3.1 Experimental Setup

3.2 Comparisons

3.3 Ablation Study

4 Limitation

L2MNet: Enhancing Continual Semantic Segmentation with Mask Matching

3.1 Preliminaries and Revisiting

3.2 Proposed Learn-to-Match Framework

3.3 Training Loss

4.1 Experimental Setting

4.2 Quantitative Evaluation

Adaptive Channel Pruning for Trainability Protection

3.1 Method Framework and Motivation

3.2 Channel Similarity Calculation and Trainability Preservation

3.3 Sparse Control and Optimization

4.1 Experiments Settings and Evaluation Metrics

4.2 Results on Imagenet

4.3 Results on Cifar-10

4.4 Results on YOLOX-s

4.5 Ablation

Exploiting Adaptive Crop and Deformable Convolution for Road Damage Detection

3 Methods

3.1 Adaptive Image Cropping Based on Vanishing Point Estimation

3.2 Feature Learning with Deformable Convolution

3.3 Diagonal Intersection over Union Loss Function

4 Experiment

4.1 Comparative Analysis of Different Datasets

4.2 Ablation Analysis

Cascaded-Scoring Tracklet Matching for Multi-object Tracking.

1 Introduction

2.1 Tracking by Detection

2.2 Joint Detection and Tracking

3.1 Cascaded-Scoring Tracklet Matching

3.2 Motion-Guided Based Target Aware

3.3 Appearance-Assisted Feature Warper

4.2 Ablation Studies

4.3 Comparison with State-of-the-Art Methods

Boosting Generalization Performance in Person Re-identification

2.1 Generalizable Person ReID

2.2 Vision-Language Learning

3.1 Review of CLIP

3.2 A Novel Cross-Modal Framework

3.3 Prompt Design Process

4.1 Datasets and Evaluation Protocols

4.4 Comparison with State-of-the-Art Methods

4.5 Other Analysis

Self-guided Transformer for Video Super-Resolution

2.1 Video Super-Resolution

2.2 Vision Transformers

3 Our Method

3.1 Network Overview

3.2 Multi-headed Self-attention Module Based on Offset-Guided Window (OGW-MSA)

3.3 Feature Aggregation (FA)

4.1 Datasets and Experimental Settings

4.2 Comparisons with State-of-the-Art Methods

SAMP: Sub-task Aware Model Pruning with Layer-Wise Channel Balancing for Person Search

3 The Proposed Method

3.1 Framework Overview

3.2 Sub-task Aware Channel Importance Estimation

3.3 Layer-Wise Channel Balancing

3.4 Adaptive OIM Loss for Model Pruning and Finetuning

4 Experimental Results and Analysis

4.1 Dataset and Evaluation Metric

4.3 Comparison with the State-of-the-Art Approaches

4.4 Ablation Study

5 Conclusion.

References

MKB: Multi-Kernel Bures Metric for Nighttime Aerial Tracking

2 Methodology

2.1 Kernel Bures Metric

2.2 Multi-Kernel Bures Metric

2.3 Objective Loss

3.1 Implementation Details

3.2 Evaluation Datasets

3.3 Comparison Results

3.4 Visualization

3.5 Ablation Study

Deep Arbitrary-Scale Unfolding Network for Color-Guided Depth Map Super-Resolution

2 The Proposed Method

2.2 Algorithm Unfolding

2.3 Continuous Up-Sampling Fusion (CUSF)

2.4 Loss Function

3 Experimental Results

3.2 The Quality Comparison of Different DSR Methods

SSDD-Net: A Lightweight and Efficient Deep Learning Model for Steel Surface Defect Detection

2.1 LMFE: Light Multiscale Feature Extraction Module

2.2 SEFF: Simple Effective Feature Fusion Network

2.3 SSDD-Net

3 Experiments and Analysis

3.2 Evaluation Metrics

3.3 Dataset

3.4 Ablation Studies

3.5 Comparison with Other SOTA Methods

3.6 Comprehensive Performance of SSDD-Net

Effective Small Ship Detection with Enhanced-YOLOv7

2.1 Small Object-Aware Feature Extraction Module (SOAFE)

2.2 Small Object-Friendly Scale-Insensitive Regression Scheme (SOFSIR)

2.3 Geometric Constraint-Based Non-Maximum Suppression Method (GCNMS)

3.1 Experimental Settings

3.2 Quantitative Analysis

3.3 Ablation Studies

3.4 Qualitative Analysis

PiDiNeXt: An Efficient Edge Detector Based on Parallel Pixel Difference Networks

2 Related Work.

2.1 The Development of Deep Learning Based Edge Detection.

Notes:

Description based on publisher supplied metadata and other sources.

Other Format:

Print version: Liu, Qingshan Pattern Recognition and Computer Vision

ISBN:

9789819985494

9819985498

OCLC:

1416189525