What is it about?
Face Template Protection Using Deep LDPC Codes Learning (DLCL) is an effective method to protect the original facials while keeping the high performance. This method maps the facials into low-density parity-check (LDPC) codes by deep CNN based multi-label learning. The simulation results on PIE and extended Yale B indicate that the proposed scheme achieves high genuine accept rate at 1% false accept rate.
Featured Image
Why is it important?
The four significant properties [28] for an ideal biometric template protection scheme: diversity, revocability, security and performance are all considered by the proposed method. Thus, the proposed scheme is robust to intra-variations without extra auxiliary data, resulting in high accuracy and security.
Perspectives
It's a great pleasure for me and co-authors to write this paper together. I hope the results of my research will enlighten other researchers. There will be more and more fantastic research works about biometric template protection.
Lingying Chen
Wuhan University of Technology
Read the Original
This page is a summary of: Face Template Protection Using Deep LDPC Codes Learning, IET Biometrics, November 2018, the Institution of Engineering and Technology (the IET),
DOI: 10.1049/iet-bmt.2018.5156.
You can read the full text:
Resources
Contributors
The following have contributed to this page
