Intro
Preface
Organization
Contents - Part II
Vision-Based Human Robot Interaction and Application
Object Tracking Algorithm Based on Dual Layer Attention
1 Introduction
2 Related Work
2.1 Visual Object Tracking
2.2 Siamese Tracking and Transformer Tracking
2.3 Transformer Tracking
3 Method
3.1 Feature Extraction
3.2 Feature Fusion Network
3.3 Prediction Head Network
3.4 Training Loss
4 Experiments
4.1 Implementation Details
4.2 Evaluation on Datasets
4.3 Ablation Study
5 Conclusions
References

Realtime 3D Reconstruction at Scale and Object Pose Estimation for Bin Picking System
1 Introduction
2 Realtime Structured Light Reconstruction
2.1 GPU's and Parallel Processing Pipelines
2.2 3D Reconstruction Algorithm Performance Test
3 Object Pose Estimation Based on PPF
3.1 Weighted Hough Voting Scheme
3.2 Postprocessing
4 Experimental Results and Analysis
4.1 Data Set Object Recognition Results
4.2 Actual Scene Reconstruction and Recognition Results
References
Large-Parallax Multi-camera Calibration Method for Indoor Wide-Baseline Scenes
1 Introduction

2 Related Work
3 Method
3.1 Building Indoor 3D Models
3.2 Direction Estimation
3.3 Translation Estimation
4 Experiment
4.1 Datasets
4.2 Comparison with Baselines
5 Conclusion
References
A Real-Time and Globally Consistent Meshing Reconstruction Without GPU
1 Introduction
2 Related Works
3 Methodology
3.1 System Overview
3.2 Depth Image Optimization
3.3 Feature Extraction and Matching Using SIMD
3.4 Robust Localization and Optimization
3.5 Online Meshing Reconstruction
4 Experiments
4.1 Comparison of Feature Extraction Time Consuming

4.2 Surface Reconstruction
4.3 Accuracy of Surface Reconstruction
5 Conclusion
References
All-in-One Image Dehazing Based on Attention Mechanism
1 Introduction
2 Related Work
2.1 Single Image Dehazing
2.2 Attentional Mechanisms
2.3 Perceived Loss
3 Our Approach
3.1 Overall Network Structure
3.2 Multi-scale Attention Mechanism
3.3 Loss Function
4 Experiment
4.1 Experiment Setup
4.2 Experimental Results and Analysis
5 Conclusion
References
Reliable AI on Machine Human Reactions

A Feature Fusion Network for Skeleton-Based Gesture Recognition
1 Introduction
2 Related Work
3 DoubleC3D Method
3.1 Overview
3.2 Generate 3D Heat Map Volumes
3.3 3D-CNN for Skeleton-Based Gesture Recognition
4 Experiments
4.1 Datasets
4.2 Implementation Details
4.3 Comparison with Previous Methods
4.4 Ablation Studies
5 Conclusion
References
Dynamic Hand Gesture Recognition Based on Multi-skeletal Features for Sign Language Recognition System
1 Introduction
2 Related Work
2.1 Video-Based Dynamic Hand Gesture Recognition