Spoofing Attack Detection in Face Recognition System Using Vision Transformer with Patch-wise Data Augmentation
Kota Watanabe, Koichi Ito, Takafumi Aoki · 2022 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) · 2022
Spoofing attacks are a serious threat to face recognition systems by malicious third parties since face images can be easily collected from the Internet. In this paper, we propose a spoofing attack detection method using Vision Transformer (ViT), which extracts features based on patches to extract fine features in a face image. We also propose a patch-wise data augmentation to improve the detection accuracy of spoofing attacks. We demonstrate the effectiveness of the proposed method through accuracy evaluation experiments using public datasets.