Abstract | ||
---|---|---|
AbstractIt has been shown that face images can be reconstructed from their representations (templates). We propose a randomized CNN to generate protected face biometric templates given the input face image and a user-specific key. The use of user-specific keys introduces randomness to the secure template and hence strengthens the template security. To further enhance the security of the templates, instead of storing the key, we store a secure sketch that can be decoded to generate the key with genuine queries submitted to the system. We have evaluated the proposed protected template generation method using three benchmarking datasets for the face (FRGC v2.0, CFP, and IJB-A). The experimental results justify that the protected template generated by the proposed method are non-invertible and cancellable, while preserving the verification performance. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/TIFS.2020.3009590 | Periodicals |
Keywords | DocType | Volume |
Bioinformatics, Face, Feature extraction, Data mining, Cryptography, Image reconstruction, Biometric, template security, deep templates, template protection, randomized CNN, protected templates | Journal | 16 |
Issue | ISSN | Citations |
1 | 1556-6013 | 3 |
PageRank | References | Authors |
0.36 | 41 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Guangcan Mai | 1 | 20 | 1.99 |
Kai Cao | 2 | 207 | 18.68 |
Lan Xiangyuan | 3 | 599 | 22.53 |
Pong C. Yuen | 4 | 2625 | 171.07 |