计算机与现代化Issue(7)：1-8,14,9.DOI:10.3969/j.issn.1006-2475.2025.07.001
基于领域生成和对比学习的人脸活体检测
Face Anti-spoofing Based on Domain Synthesis and Contrastive Learning
摘要
Abstract
Face anti-spoofing(FAS)is an important mean to guarantee the security of face recognition systems.Existing FAS methods have poor generalization in cross-dataset testing scenarios,which leads to a drastic performance degradation.To this way,this paper proposes a FAS method based on domain synthesis and contrastive learning.The proposed method mainly con-tains two novel modules:domain synthesis module and contrastive learning module.The former randomly swaps local regions of face images in different source domains at the image level to generate pseudo-source domain samples.Then,the above face im-ages are reconstructed at the feature-level by exchanging the local features of the corresponding positions reconstructed at the image-level.By maximizing the similarity of the reconstructed sample's features and the reconstructed features,this module can ensure the stability of the generated pseudo source domain while expanding the number of samples and attack types.This pro-vides a solid data foundation for the proposed method to learn generalized feature spaces.The later minimizes the intra-class dis-tance of the real face representation and maximizes the inter-class distance between the real face and spoof faces.Meanwhile,this module maximizes the inter-class distance between the real face and the reconstructed samples.This process effectively pro-motes intra-class compactness of the real face and ensures that the proposed method can learn a good decision curve.The pro-posed method is trained and tested on four publicly available face live detection datasets CASIA-FASD,Replay-Attack,MSU-MFSD,and OULU-NPU,and the experimental results show that the proposed method has a good generalization performance in cross-dataset testing scenarios.关键词
人脸活体检测/伪源域/对比学习/领域泛化Key words
face anti-spoofing/pseudo source domain/contrastive learning/domain generalization分类
信息技术与安全科学引用本文复制引用
张山鹿,张伟..基于领域生成和对比学习的人脸活体检测[J].计算机与现代化,2025,(7):1-8,14,9.基金项目
西安工业大学研究生创新实践项目(2022030615) （2022030615）
