The Deep Neural Network Approach (with Data and Code)
Face recognition is the most prominent biometric technique for identity authentication and is widely used in applications, such as access control, finance, law enforcement, and public security. It is a long-standing research, with over 50 years of research and developments. Advances in deep learning and neural networks have significantly reshaped the landscape of face recognition research and applications in almost all aspects since the first two editions (1st Edition and 2nd Edition) of this Handbook.
This third edition, while inheriting the title of Handbook of Face Recognition, will be composed of entirely new contents describing the latest face recognition methodologies and technologies in the deep neural network framework. The book presents a unified resource of theory, algorithms and implementations to bring students, researchers and practitioners to all aspects of face recognition. The book not only presents the latest developments in methods and algorithms (Parts I through IV), but also provides data and code to allow for hands-on learning and developing reproducible face recognition algorithms and systems (Appendix) by deep learning programming. The data and code will be released at Github and will be updated subsequently to keep the materials up to date.
We will invite senior researchers to review their assigned topics, illustrate the state-of-the-art methods and give an expectation of future development. The estimated time for completing the book is 12 months (from Jan 2020 to Dec 2020).
The open source part will cover the most straightforward and effective methods on deep face recognition. All of the code will be passed tests under the newest Pytorch version.
We appreciate all suggestions (more suitable content and arrangement) and contributions (data annotation and code optimization) to improve this book.
The book will be released under the license of Creative Commons Attribution 4.0 International (CC BY 4.0).
The open source code will be released under the license of Apache 2.0.