Towards Generalizable and Robust Face Manipulation Detection via Bag-of-local-feature

• DOI: 10.48550/arxiv.2103.07915
• Publication Date: 2021-01-01
• Authors: Miao, Changtao; Chu, Qi; Li, Weihai; Gong, Tao; Zhuang, Wanyi; Yu, Nenghai
• Keywords: FOS: Computer and information sciences; Computer Vision and Pattern Recognition (cs.CV); Computer Science - Computer Vision and Pattern Recognition; 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

5 pages, 2 figures<br/>Over the past several years, in order to solve the problem of malicious abuse of facial manipulation technology, face manipulation detection technology has obtained considerable attention and achieved remarkable progress. However, most existing methods have very impoverished generalization ability and robustness. In this paper, we propose a novel method for face manipulation detection, which can improve the generalization ability and robustness by bag-of-local-feature. Specifically, we extend Transformers using bag-of-feature approach to encode inter-patch relationships, allowing it to learn local forgery features without any explicit supervision. Extensive experiments demonstrate that our method can outperform competing state-of-the-art methods on FaceForensics++, Celeb-DF and DeeperForensics-1.0 datasets.<br/>