2020/11/16 by Anjith George, Sébastien Marcel, George, Anjith +1 · 2 citations
Computer Science · #Biometric Identification and Security #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Face recognition and analysis
paper · pdf · doi:10.48550/arxiv.2011.08019
openalex publication_date 2020/11/16 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
The vulnerability of face recognition systems to presentation attacks has\nlimited their application in security-critical scenarios. Automatic methods of\ndetecting such malicious attempts are essential for the safe use of facial\nrecognition technology. Although various methods have been suggested for\ndetecting such attacks, most of them over-fit the training set and fail in\ngeneralizing to unseen attacks and environments. In this work, we use transfer\nlearning from the vision transformer model for the zero-shot anti-spoofing\ntask. The effectiveness of the proposed approach is demonstrated through\nexperiments in publicly available datasets. The proposed approach outperforms\nthe state-of-the-art methods in the zero-shot protocols in the HQ-WMCA and\nSiW-M datasets by a large margin. Besides, the model achieves a significant\nboost in cross-database performance as well.\n