Vision-Based Human Activity Recognition

This book provides a comprehensive review of V-HAR, covering tasks, technologies, and applications, with a focus on deep learning-based approaches. It discusses various human activities, sensors, and vision-based methods, and highlights the importance of V-HAR in various fields.

Format: Paperback / softback
Length: 121 pages
Publication date: 23 April 2022
Publisher: Springer Verlag, Singapore

The field of Human Activity Recognition (HAR) has emerged as a highly captivating and rapidly advancing research domain, driven by the proliferation of diverse sensors, real-time data streaming, and remarkable advancements in computer vision, machine learning, and related fields. HAR has immense potential applications across a wide range of scenarios, including medical diagnosis, video surveillance, public governance, and human-machine interaction. It involves the identification and classification of various human activities, such as walking, running, sitting, sleeping, standing, showering, cooking, driving, and abnormal behaviors.

Data for HAR can be obtained from wearable sensors, accelerometers, video frames, or images. Among these sensors, vision-based sensors have gained widespread popularity due to their cost-effectiveness, high-quality, and non-intrusive nature. Consequently, vision-based human activity recognition (V-HAR) has emerged as the most significant and widely adopted category within the broader realm of HAR technologies.

This comprehensive book offers a systematic, comprehensive, and timely review on V-HAR, encompassing the related tasks, cutting-edge technologies, and applications of V-HAR, particularly focusing on deep learning-based approaches. It delves into the latest advancements and commonly used benchmark in the field, providing valuable insights for both researchers and practitioners. Furthermore, the book explores future directions and recommendations for emerging researchers to further contribute to the growth and development of V-HAR.

The book is organized into several chapters, each dedicated to exploring different aspects of V-HAR. The first chapter provides an introduction to the topic, highlighting its significance and potential applications. It also provides a brief overview of the historical development and the current state-of-the-art in V-HAR.

The subsequent chapters delve into specific topics related to V-HAR. Chapter 2 discusses hand gestures, exploring various techniques for their recognition, including feature extraction, classification, and evaluation. Chapter 3 focuses on head pose estimation, discussing the challenges and recent advancements in capturing and analyzing head movements. Chapter 4 explores body activity recognition, discussing the use of wearable sensors and machine learning algorithms to detect and classify human activities. Chapter 5 explores eye gaze estimation, discussing the challenges of capturing and analyzing eye movements for HAR. Chapter 6 discusses attention modeling, exploring the use of deep learning techniques to understand human attention and its relationship with activities.

In addition to the technical discussions, the book also includes case studies and real-world applications to demonstrate the practical implications of V-HAR. These case studies showcase the successful implementation of V-HAR in various scenarios, such as healthcare, surveillance, and human-computer interaction.

Overall, this book serves as a valuable resource for researchers, practitioners, and students interested in V-HAR. It provides a comprehensive coverage of the field, covering the latest advancements, commonly used benchmark, and future directions. With its extensive content and practical applications, the book will be of great interest to anyone seeking to explore the potential of vision-based human activity recognition in various domains.

Weight: 215g
Dimension: 235 x 155 (mm)
ISBN-13: 9789811922893
Edition number: 1st ed. 2022