Springer, 2025. — 213 p. — ISBN 978-981-97-9333-4.
Анализ позы человека: глубокое обучение соответствует кинематике человека в видео.
With the rapid technological advancements and the proliferation of digital media in the contemporary world, the study of human behavior using computer technologies has taken on new dimensions. Among the most intriguing and consequential areas of research within this field is the analysis of human pose in video. Understanding human behavior in video offers insights into the complexities of human nature, social dynamics, cognitive processes, and human-machine interactions, with wide applications in fields such as security, surveillance, sports analysis, health-care monitoring, and entertainment. It provides a rich source of information for addressing societal challenges, improving human-computer interaction, and enhancing our understanding of ourselves and the world around us.
This book stands at the intersection of computer vision, artificial intelligence, and human kinematics, offering a comprehensive exploration of the principles, methodologies, and applications of human pose analysis in video data. The concept of human pose analysis has two contents in this book: human body pose analysis and human head pose analysis. In specific, human body pose analysis involves estimating/predicting the position and orientation of all the major joints and body parts of a person, such as the head, neck, shoulders, elbows, wrists, hips, knees, and ankles. It aims to capture the full body posture in a 2D or 3D space. Human head pose analysis is more specific and focuses only on the orientation of the head.
It typically estimates/predicts the angles of rotation around the yaw, pitch, and roll axes to understand the direction in which a person is looking or tilting his/her head.
IntroductionPart I 2D Human Body Pose Analysis
Bidirectionally Learning Heatmaps for 2D Human Pose Estimation
Self-Supervised Multi-person 2D Human Pose Estimation
Part II 3D Human Body Pose Analysis
Bidirectional 2D-3D Transformation for 3D Human Pose Estimation
Joint Data Augmentation and Representation for 3D Human Pose Estimation
Spatial-temporal Feature Transform for 3D Human Pose Estimation
Real-Time 3D Human Pose Estimation from a Single RGB Image
Part III 3D Human Head Pose Analysis
Spatial-Temporal Aggregation for 3D Human Head Pose Estimation
Spatial-Temporal Pyramid for 3D Human Head Pose Prediction
Conclusion and Outlook