Linked e-resources
Details
Table of Contents
Preface; Outline of the Book and Chapter Synopsis; Challenges for the Future; Acknowledgements; Contents; List of Figures; List of Tables; Deep Learning for Face Biometrics; 1 The Functional Neuroanatomy of Face Processing: Insights from Neuroimaging and Implications for Deep Learning; 1.1 The Functional Characteristics and Organization of the Ventral Face Network in the Human Brain; 1.1.1 Functional Characteristics of the Ventral Face Network; 1.2 The Neural Architecture and Connections of the Ventral Face Network
1.2.1 The Functional Organization of the Face Network Is Consistent Across Participants1.2.2 The Cytoarchitecture of Face-Selective Regions; 1.2.3 White Matter Connections of the Ventral Face Network; 1.3 Computations by Population Receptive Fields in the Ventral Face Network; 1.3.1 pRF Measurements Reveal a Hierarchical Organization of the Face Network; 1.3.2 Attention Modulates pRF Properties, Enhancing Peripheral Representations Where Visual Acuity Is the Worst; 1.4 Eyes to the Future: Computational Insights from Anatomical and Functional Features of the Face Network
1.4.1 What Is the Computational Utility of the Organized Structure of the Cortical Face Network?1.4.2 What Can Deep Convolutional Networks Inform About Computational Strategies of the Brain?; 1.5 Conclusions; References; 2 Real-Time Face Identification via Multi-convolutional Neural Network and Boosted Hashing Forest; 2.1 Introduction; 2.2 Related Work; 2.3 CNHF with Multiple Convolution CNN; 2.4 Learning Face Representation via Boosted Hashing Forest; 2.4.1 Boosted SSC, Forest Hashing and Boosted Hashing Forest; 2.4.2 BHF: Objective-Driven Recurrent Coding
2.4.3 BHF: Learning Elementary Projection via RANSAC Algorithm2.4.4 BHF: Boosted Hashing Forest; 2.4.5 BHF: Hashing Forest as a Metric Space; 2.4.6 BHF: Objective Function for Face Verification and Identification; 2.4.7 BHF Implementation for Learning Face Representation; 2.5 Experiments; 2.5.1 Methodology: Learning and Testing CNHF; 2.5.2 Hamming Embedding: CNHL Versus CNN, BHF Versus Boosted SSC; 2.5.3 CNHF: Performance w.r.t. Depth of Trees; 2.5.4 CNHL and CNHF Versus Best Methods on LFW; 2.6 Conclusion and Discussion; References
3 CMS-RCNN: Contextual Multi-Scale Region-Based CNN for Unconstrained Face Detection3.1 Introduction; 3.2 Related Work; 3.3 Background in Deep Convolution Nets; 3.3.1 Region-Based Convolution Neural Networks; 3.3.2 Limitations of Faster R-CNN; 3.3.3 Other Face Detection Method Limitations; 3.4 Contextual Multi-Scale R-CNN; 3.4.1 Identifying Tiny Faces; 3.4.2 Integrating Body Context; 3.4.3 Information Fusion; 3.4.4 Implementation Details; 3.5 Experiments; 3.5.1 Experiments on WIDER FACE Dataset; 3.5.2 Experiments on FDDB Face Database; 3.6 Conclusion and Future Work; References; Deep Learning for Fingerprint, Fingervein and Iris Recognition.
1.2.1 The Functional Organization of the Face Network Is Consistent Across Participants1.2.2 The Cytoarchitecture of Face-Selective Regions; 1.2.3 White Matter Connections of the Ventral Face Network; 1.3 Computations by Population Receptive Fields in the Ventral Face Network; 1.3.1 pRF Measurements Reveal a Hierarchical Organization of the Face Network; 1.3.2 Attention Modulates pRF Properties, Enhancing Peripheral Representations Where Visual Acuity Is the Worst; 1.4 Eyes to the Future: Computational Insights from Anatomical and Functional Features of the Face Network
1.4.1 What Is the Computational Utility of the Organized Structure of the Cortical Face Network?1.4.2 What Can Deep Convolutional Networks Inform About Computational Strategies of the Brain?; 1.5 Conclusions; References; 2 Real-Time Face Identification via Multi-convolutional Neural Network and Boosted Hashing Forest; 2.1 Introduction; 2.2 Related Work; 2.3 CNHF with Multiple Convolution CNN; 2.4 Learning Face Representation via Boosted Hashing Forest; 2.4.1 Boosted SSC, Forest Hashing and Boosted Hashing Forest; 2.4.2 BHF: Objective-Driven Recurrent Coding
2.4.3 BHF: Learning Elementary Projection via RANSAC Algorithm2.4.4 BHF: Boosted Hashing Forest; 2.4.5 BHF: Hashing Forest as a Metric Space; 2.4.6 BHF: Objective Function for Face Verification and Identification; 2.4.7 BHF Implementation for Learning Face Representation; 2.5 Experiments; 2.5.1 Methodology: Learning and Testing CNHF; 2.5.2 Hamming Embedding: CNHL Versus CNN, BHF Versus Boosted SSC; 2.5.3 CNHF: Performance w.r.t. Depth of Trees; 2.5.4 CNHL and CNHF Versus Best Methods on LFW; 2.6 Conclusion and Discussion; References
3 CMS-RCNN: Contextual Multi-Scale Region-Based CNN for Unconstrained Face Detection3.1 Introduction; 3.2 Related Work; 3.3 Background in Deep Convolution Nets; 3.3.1 Region-Based Convolution Neural Networks; 3.3.2 Limitations of Faster R-CNN; 3.3.3 Other Face Detection Method Limitations; 3.4 Contextual Multi-Scale R-CNN; 3.4.1 Identifying Tiny Faces; 3.4.2 Integrating Body Context; 3.4.3 Information Fusion; 3.4.4 Implementation Details; 3.5 Experiments; 3.5.1 Experiments on WIDER FACE Dataset; 3.5.2 Experiments on FDDB Face Database; 3.6 Conclusion and Future Work; References; Deep Learning for Fingerprint, Fingervein and Iris Recognition.