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

Format:

Book

Contributor:

Liu, Qingshan, Editor.

Wang, Hanzi., Editor.

Ma, Zhanyu., Editor.

Zheng, Weishi., Editor.

Zha, Hongbin, Editor.

Chen, Xilin, Editor.

Wang, Liang, Editor.

Ji, Rongrong, Editor.

Series:

Lecture Notes in Computer Science, 1611-3349 ; 14432

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 (XIV, 513 p. 157 illus., 152 illus. in color.)

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 VIII

Neural Network and Deep Learning I

A Quantum-Based Attention Mechanism in Scene Text Detection

1 Introduction

2 Related Work

2.1 Attention Mechanism

2.2 Revisit Quantum-State-based Mapping

3 Approach

3.1 QSM-Based Channel Attention (QCA) Module and QSM-Based Spatial Attention (QSA) Module

3.2 Quantum-Based Convolutional Attention Module (QCAM)

3.3 Adaptive Channel Information Transfer Module (ACTM)

4 Experiments

4.1 Implementation Details

4.2 Performance Comparison

4.3 Ablation Study

5 Discussion and Conclusion

References

NCMatch: Semi-supervised Learning with Noisy Labels via Noisy Sample Filter and Contrastive Learning

2.1 Semi-supervised Learning

2.2 Self-supervised Contrastive Learning

2.3 Learning with Noisy Labels

3 Method

3.1 Preliminaries

3.2 Overall Framework

3.3 Noisy Sample Filter (NSF)

3.4 Semi-supervised Contrastive Learning (SSCL)

4.1 Datasets

4.2 Experimental for SSL

4.3 Experimental for SSLNL

4.4 Ablation Study

5 Conclusion

Data-Free Low-Bit Quantization via Dynamic Multi-teacher Knowledge Distillation

3.2 More Insight on 8-Bit Quantized Models

3.3 Dynamic Multi-teacher Knowledge Distillation

4.1 Experimental Setups

4.2 Comparison with Previous Data-Free Quantization Methods

4.3 Ablation Studies

LeViT-UNet: Make Faster Encoders with Transformer for Medical Image Segmentation

2 Related Works

3.1 Architecture of LeViT-UNet

3.2 LeViT as Encoder

3.3 CNNs as Decoder

4 Experiments and Results

4.1 Dataset

4.2 Implementation Details.

4.3 Experiment Results on Synapse Dataset

4.4 Experiment Results on ACDC Dataset

DUFormer: Solving Power Line Detection Task in Aerial Images Using Semantic Segmentation

2.1 Vision Transformer

2.2 Semantic Segmentation

3 Proposed Architecture

3.1 Overview

3.2 Double U Block (DUB)

3.3 Power Line Aware Block (PLAB)

3.4 BiscSE Block

3.5 Loss Function

4.1 Experimental Settings

4.2 Comparative Experiments

4.3 Ablation Experiments

Space-Transform Margin Loss with Mixup for Long-Tailed Visual Recognition

2.1 Mixup and Its Space Transformation

2.2 Long-Tailed Learning with Mixup

2.3 Re-balanced Loss Function Modification Methods

3.1 Space Transformation in Mixup

3.2 Space-Transform Margin Loss Function

4.2 Implementations Details

4.3 Main Results

4.4 Feature Visualization and Analysis of STM Loss

4.5 Ablation Study

A Multi-perspective Squeeze Excitation Classifier Based on Vision Transformer for Few Shot Image Classification

3.1 Problem Definition

3.2 Meta-Training Phase

3.3 Meta-test Phase

4 Experimental Results

4.1 Datasets and Training Details

4.2 Evaluation Results

ITCNN: Incremental Learning Network Based on ITDA and Tree Hierarchical CNN

2 Proposed Network

2.1 Network Structure

2.2 ITDA

2.3 Branch Route

2.4 Training Strategies

2.5 Optimization Strategies

3 Experiments and Results

3.1 Experiment on Classification

3.2 Experiment on CIL

4 Conclusion

References.

Periodic-Aware Network for Fine-Grained Action Recognition

2.1 Skeleton-Based Action Recognition

2.2 Periodicity Estimation of Videos

2.3 Squeeze and Excitation Module

3.1 3D-CNN Backbone

3.2 Periodicity Feature Extraction Module

3.3 Periodicity Fusion Module

4 Experiment

4.2 Implementation Details

4.4 Comparison with State-of-the-Art Methods

Learning Domain-Invariant Representations from Text for Domain Generalization

2.1 Domain Generalization

2.2 CLIP in Domain Generalization

3.1 Problem Formulation

3.2 Text Regularization

3.3 CLIP Representations

4.1 Datasets and Experimental Settings

4.2 Comparison with Existing DG Methods

5 Conclusions

TSTD:A Cross-modal Two Stages Network with New Trans-decoder for Point Cloud Semantic Segmentation

2.1 Image Transformers

2.2 Point Cloud Transformer

2.3 Joint 2D-3D Network

3.1 Overall Architecture

3.2 2D-3D Backprojection

3.3 Trans-Decoder

4.1 Dataset and Metric

4.3 Ablation Experiment

NeuralMAE: Data-Efficient Neural Architecture Predictor with Masked Autoencoder

2.1 Neural Architecture Performance Predictors

2.2 Generative Self-supervised Learning

3.1 Overall Framework

3.2 Pre-training

3.3 Fine-Tuning

3.4 Multi-head Attention-Masked Transformer

4.2 Experiments on NAS-Bench-101

4.3 Experiments on NAS-Bench-201

4.4 Experiments on NAS-Bench-301.

4.5 Ablation Study

Co-regularized Facial Age Estimation with Graph-Causal Learning

2 Method

2.1 Problem Formulation

2.2 Ordinal Decision Mapping

2.3 Bilateral Counterfactual Pooling

3 Experiments

3.1 Datasets and Evaluation Settings

3.2 Comparison with State-of-the-Art Methods

3.3 Ablation Study

3.4 Performance Under Out-of-Distribution Settings

3.5 Qualitative Results

Online Distillation and Preferences Fusion for Graph Convolutional Network-Based Sequential Recommendation

2.1 Graph Construction

2.2 Collaborative Learning

2.3 Feature Fusion

3 Experiment

3.1 Experimental Setup

3.2 Experimental Results

3.3 Ablation Studies

Grassmann Graph Embedding for Few-Shot Class Incremental Learning

3 The Proposed Method

3.2 Overview

3.3 Grassmann Manifold Embedding

3.4 Graph Structure Preserving on Grassmann Manifold

4.1 Experimental Setup

4.2 Comparison with State-of-the-Art Methods

Global Variational Convolution Network for Semi-supervised Node Classification on Large-Scale Graphs

3 Proposed Methods

3.1 Positive Pointwise Mutual Information on Large-Scale Graphs

3.2 Global Variational Aggregation

3.3 Variational Convolution Kernels

4.1 Comparison Experiments

4.2 Ablation Study

4.3 Runtime Study

Frequency Domain Distillation for Data-Free Quantization of Vision Transformer

2.1 Vision Transformer (ViT)

2.2 Network Quantization

3 Preliminaries

3.1 Quantizer.

3.2 Fast Fourier Transform (FFT) and Frequency Domain

4 Method

4.1 Our Insights

4.2 Frequency Domain Distillation

4.3 The Overall Pipeline

5 Experimentation

5.1 Comparison Experiments

5.2 Ablation Study

6 Conclusions

An ANN-Guided Approach to Task-Free Continual Learning with Spiking Neural Networks

2.1 Image Generation in SNNs

2.2 Continual Learning

3 Preliminary

3.1 The Referee Module: WGAN

3.2 The Player Module: FSVAE

4 Methodology

4.1 Problem Setting

4.2 Overview of Our Model

4.3 Adversarial Similarity Expansion

4.4 Precise Pruning

5 Experimental Results

5.1 Dataset Setup

5.2 Classification Tasks Under TFCL

5.3 The Impact of Different Thresholds and Buffer Sizes

5.4 ANN and SNN Under TFCL

6 Conclusion

Multi-adversarial Adaptive Transformers for Joint Multi-agent Trajectory Prediction

2.1 Multi-agent Trajectory Prediction

2.2 Domain Adaptation

3 Proposed Method

3.1 Encoder: Processing Multi-aspect Data

3.2 Decoder: Generating Multi-modal Trajectories

3.3 Adaptation: Learning Doamin Invaint Feature

3.4 Loss Function

4.2 Problem Setting

4.3 Evaluation Metrics

4.4 Implementation Details

4.5 Quantitative Analysis

4.6 Ablation Study

Enhancing Open-Set Object Detection via Uncertainty-Boxes Identification

3 Methodology

3.1 Preliminary

3.2 Baseline Setup

3.3 Pseudo Proposal Advisor

3.4 Uncertainty-Box Detection

4.2 Comparison with Other Methods

4.4 Visualization and Qualitative Analysis

Interventional Supervised Learning for Person Re-identification.

Notes:

Description based on publisher supplied metadata and other sources.

ISBN:

9789819985432

9819985439

OCLC:

1416698802