Vision-Based Human Activity Recognition (SpringerBriefs in Intelligent Systems)

✍ Scribed by Zhongxu Hu, Chen Lv

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 130
• Category: Library

✦ Synopsis

This book offers a systematic, comprehensive, and timely review on V-HAR, and it covers the related tasks, cutting-edge technologies, and applications of V-HAR, especially the deep learning-based approaches. The field of Human Activity Recognition (HAR) has become one of the trendiest research topics due to the availability of various sensors, live streaming of data and the advancement in computer vision, machine learning, etc. HAR can be extensively used in many scenarios, for example, medical diagnosis, video surveillance, public governance, also in human–machine interaction applications. In HAR, various human activities such as walking, running, sitting, sleeping, standing, showering, cooking, driving, abnormal activities, etc., are recognized. The data can be collected from wearable sensors or accelerometer or through video frames or images; among all the sensors, vision-based sensors are now the most widely used sensors due to their low-cost, high-quality, and unintrusivecharacteristics. Therefore, vision-based human activity recognition (V-HAR) is the most important and commonly used category among all HAR technologies.

The addressed topics include hand gestures, head pose, body activity, eye gaze, attention modeling, etc. The latest advancements and the commonly used benchmark are given. Furthermore, this book also discusses the future directions and recommendations for the new researchers.

✦ Table of Contents

Preface
Contents
Acronyms
1 Introduction
1.1 Background of Human Activity Recognition
1.2 Commonly Used Sensor Types
1.3 Taxonomy of Vision-Based Human Activity Recognition
1.4 Summary
References
2 Vision-Based Hand Activity Recognition
2.1 Introduction
2.1.1 Hand Recognition with Marker
2.1.2 Hand Recognition Without Marker
2.1.3 Summary
2.2 Depth Sensor-Based Hand Pose Estimation
2.2.1 Neural Network Basics
2.2.2 CNN Model for Hand Pose Estimation
2.2.3 Multi-scale Optimization
2.2.4 Multi-frame Optimization
2.3 Efficient Dynamic Hand Gesture Recognition
2.3.1 Pre-processing
2.3.2 3D CNN-Based Network Structure
2.3.3 Architecture Optimization
2.4 Summary
References
3 Vision-Based Facial Activity Recognition
3.1 Introduction
3.2 Appearance-Based Head Pose Estimation
3.2.1 End-To-End Head Pose Estimation Model
3.2.2 Model Analysis
3.2.3 Summary
3.3 Dynamic Head Tracking System
3.3.1 Vision-Based Driver Head Pose Tracker
3.3.2 Model Analysis
3.3.3 Summary
3.4 Appearance-Based Eye Gaze Estimation
3.4.1 Gaze Direction Estimation
3.4.2 Gaze Fixation Tracking
3.5 Summary
References
4 Vision-Based Body Activity Recognition
4.1 Introduction
4.2 Vision-Based Body Pose Estimation
4.2.1 Top-Down Methods for Pose Estimation
4.2.2 Bottom-Up Methods for Pose Estimation
4.2.3 Common Datasets
4.3 Vision-Based Action Recognition
4.3.1 Spatial–temporal-Based Action Recognition
4.3.2 Skeleton-Based Action Recognition
4.3.3 Common Datasets
4.4 Vision-Based Body Reconstruction
4.4.1 Model-Based Reconstruction
4.4.2 Fusion-Based Reconstruction
4.4.3 Neural Rendering-Based Reconstruction
4.5 Summary
References
5 Vision-Based Human Attention Modelling
5.1 Introduction
5.2 Visual Saliency Map Estimation
5.3 Context-Aware Human Attention Estimation
5.3.1 Methodology
5.3.2 Model Analysis
5.4 Summary
References
6 Conclusions and Recommendations
6.1 Conclusions
6.2 Recommendations