| Description |
1 online resource (xii, 156 pages) : illustrations (some color) |
| Contents |
Intro; Foreword; Preface; Contents; Part I Motivation and Overview; 1 The Foundation and Advances of Deep Learning; 1.1 Neural Networks; 1.1.1 Perceptron; 1.1.2 Multilayer Perceptron; 1.1.3 Formulation of Neural Network; 1.2 New Techniques in Deep Learning; 1.2.1 Batch Normalization; 1.2.2 Batch Kalman Normalization; References; 2 Human-Centric Visual Analysis: Tasks and Progress; 2.1 Face Detection; 2.2 Facial Landmark Localization; 2.2.1 Conventional Approaches; 2.2.2 Deep-Learning-Based Models; 2.3 Pedestrian Detection; 2.3.1 Benchmarks for Pedestrian Detection |
|
2.3.2 Pedestrian Detection Methods2.4 Human Segmentation and Clothes Parsing; References; Part II Localizing Persons in Images; 3 Face Localization and Enhancement; 3.1 Facial Landmark Machines; 3.2 The Cascaded BB-FCN Architecture; 3.2.1 Backbone Network; 3.2.2 Branch Network; 3.2.3 Ground Truth Heat Map Generation; 3.3 Experimental Results; 3.3.1 Datasets; 3.3.2 Evaluation Metric; 3.3.3 Performance Evaluation for Unconstrained Settings; 3.3.4 Comparison with the State of the Art; 3.4 Attention-Aware Face Hallucination; 3.4.1 The Framework of Attention-Aware Face Hallucination |
|
3.4.2 Recurrent Policy Network3.4.3 Local Enhancement Network; 3.4.4 Deep Reinforcement Learning; 3.4.5 Experiments; References; 4 Pedestrian Detection with RPN and Boosted Forest; 4.1 Introduction; 4.2 Approach; 4.2.1 Region Proposal Network for Pedestrian Detection; 4.2.2 Feature Extraction; 4.2.3 Boosted Forest; 4.3 Experiments and Analysis; References; Part III Parsing Person in Detail; 5 Self-supervised Structure-Sensitive Learning for Human Parsing; 5.1 Introduction; 5.2 Look into Person Benchmark; 5.3 Self-supervised Structure-Sensitive Learning |
|
5.3.1 Self-supervised Structure-Sensitive Loss5.3.2 Experimental Result; References; 6 Instance-Level Human Parsing; 6.1 Introduction; 6.2 Related Work; 6.3 Crowd Instance-Level Human Parsing Dataset; 6.3.1 Image Annotation; 6.3.2 Dataset Statistics; 6.4 Part Grouping Network; 6.4.1 PGN Architecture; 6.4.2 Instance Partition Process; 6.5 Experiments; 6.5.1 Experimental Settings; 6.5.2 PASCAL-Person-Part Dataset; 6.5.3 CIHP Dataset; 6.5.4 Qualitative Results; References; 7 Video Instance-Level Human Parsing; 7.1 Introduction; 7.2 Video Instance-Level Parsing Dataset |
|
7.2.1 Data Amount and Quality7.2.2 Dataset Statistics; 7.3 Adaptive Temporal Encoding Network; 7.3.1 Flow-Guided Feature Propagation; 7.3.2 Parsing R-CNN; 7.3.3 Training and Inference; References; Part IV Identifying and Verifying Persons; 8 Person Verification; 8.1 Introduction; 8.2 Generalized Similarity Measures; 8.2.1 Model Formulation; 8.2.2 Connection with Existing Models; 8.3 Joint Similarity and Feature Learning; 8.3.1 Deep Architecture; 8.3.2 Model Training; 8.4 Experiments; References; 9 Face Verification; 9.1 Introduction; 9.2 Related Work; 9.3 Framework Overview |
| Summary |
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 |
| Bibliography |
Includes bibliographical references |
| Notes |
Online resource; title from PDF title page (SpringerLink, viewed November 18, 2019) |
| Subject |
Computer vision.
|
|
Optical pattern recognition.
|
|
Machine learning.
|
|
Computer vision
|
|
Machine learning
|
|
Optical pattern recognition
|
| Form |
Electronic book
|
| Author |
Zhang, Dongyu, author
|
|
Luo, Ping, author
|
|
Zuo, Wangmeng, author.
|
| ISBN |
9789811323874 |
|
9811323879 |
|
