Liang Lin – författare

This book introduces the applications of deep learning in various human centric visual analysis tasks, including classical ones like face detection and alignment and some newly rising tasks like fashion clothing parsing. Starting from an overview of current research in human centric visual analysis, the book then presents a tutorial of basic concepts and techniques of deep learning. In addition, the book systematically investigates the main human centric analysis tasks of different levels, ranging from detection and segmentation to parsing and higher-level understanding. At last, it presents the state-of-the-art solutions based on deep learning for every task, as well as providing sufficient references and extensive discussions. Specifically, this book addresses four important research topics, including 1) localizing persons in images, such as face and pedestrian detection; 2) parsing persons in details, such as human pose and clothing parsing, 3) identifying and verifying persons, such as face and human identification, and 4) high-level human centric tasks, such as person attributes and human activity understanding.This book can serve as reading material and reference text for academic professors / students or industrial engineers working in the field of vision surveillance, biometrics, and human-computer interaction, where human centric visual analysis are indispensable in analysing human identity, pose, attributes, and behaviours for further understanding.

This book introduces the applications of deep learning in various human centric visual analysis tasks, including classical ones like face detection and alignment and some newly rising tasks like fashion clothing parsing. Starting from an overview of current research in human centric visual analysis, the book then presents a tutorial of basic concepts and techniques of deep learning. In addition, the book systematically investigates the main human centric analysis tasks of different levels, ranging from detection and segmentation to parsing and higher-level understanding. At last, it presents the state-of-the-art solutions based on deep learning for every task, as well as providing sufficient references and extensive discussions.

Specifically, this book addresses four important research topics, including 1) localizing persons in images, such as face and pedestrian detection; 2) parsing persons in details, such as human pose and clothing parsing, 3) identifying and verifying persons, such as face and human identification, and 4) high-level human centric tasks, such as person attributes and human activity understanding.

This book can serve as reading material and reference text for academic professors / students or industrial engineers working in the field of vision surveillance, biometrics, and human-computer interaction, where human centric visual analysis are indispensable in analysing human identity, pose, attributes, and behaviours for further understanding.

This book provides a systematic overview of the mainstream technical approaches in embodied intelligence, constructing a comprehensive knowledge framework. It offers in-depth explanations of key embodied AI technologies, including perception, navigation, manipulation, and planning, as well as multi-agent interaction, communication, and cooperation. By doing so, it establishes a structured learning architecture for researchers, engineers, and students in the fields of robotics and embodied AI.This book is divided into eight chapters, with the content organized in a progressive manner from basic to advanced concepts, aiming to build a comprehensive cognitive framework for readers in the field of embodied intelligence, covering both theoretical systems and practical methods. Chapter 2 focuses on the foundational technical system of embodied intelligence, highlighting core content such as 3D spatial representation, reinforcement learning methods, and large model technologies. These foundational technologies form the theoretical backbone of current embodied intelligence algorithms, and a deep understanding of them is a crucial prerequisite for grasping the subsequent content. Chapters 3 to 7 systematically introduce the key technical systems of embodied intelligence from five dimensions: perception, navigation, manipulation, planning, and collaboration. Chapter 8, oriented towards practice, focuses on the implementation paths of embodied intelligence in simulation platforms.The translation was done with the help of artificial intelligence. A subsequent human revision was done primarily in terms of content.

This book focuses on the new paradigm of artificial intelligence and systematically introduces the key technologies, foundational models, and typical applications of multimodal large models. To make the technical content more accessible for lower-year undergraduate students and newcomers to the AI field, the book presents each key technical point in an easy-to-understand manner and provides numerous intuitive examples. It deeply analyses the structure and technology of several classic multimodal large models. The aim is to offer readers a clear guide to the technical methods, open-source platforms, and application scenarios of multimodal large models, as well as to provide insights into achieving general artificial intelligence, including cutting-edge technologies such as causal reasoning, world models, embodied intelligence, and multi-agent systems. The book aspires to provide a clear perspective for both academia and industry, helping AI researchers gain a more comprehensive understanding of multimodal large model technologies and the development directions of the next generation of artificial intelligence.The book is divided into five chapters. Chapter 1 explores the most representative large model structures in depth. Chapter 2 provides a thorough analysis of the core technologies of multimodal large models. Chapter 3 introduces several representative multimodal large models. Chapter 4 delves into three typical applications: visual question answering, AI-generated content (AIGC), and embodied intelligence. Chapter 5 discusses feasible approaches to achieving general artificial intelligence.This book is suitable not only as a textbook for senior undergraduate and graduate students in relevant university programs but also as an essential reference for IT professionals. The Chinese version of this book has been selected for the undergraduate textbook series at Sun Yat-sen University.The translation was done with the help of artificial intelligence. A subsequent human revision was done primarily in terms of content.