论文标题
在全球范围内局部推断：一种可推广的面部抗刺激方法
Look Locally Infer Globally: A Generalizable Face Anti-Spoofing Approach
论文作者
论文摘要
最先进的欺骗检测方法倾向于过度拟合在训练过程中看到的欺骗类型，并且未能推广到未知的欺骗类型。鉴于面对反欺骗是本质上的一项本地任务，我们提出了一个面部反欺骗框架，即自我监督的区域完全卷积网络（SSR-FCN），该培训是在自助培训的方式中从脸部图像中学习局部歧视性线索的训练。所提出的框架可提高可普遍性，同时保持整体面部抗疾病方法的计算效率（在NVIDIA GTX 1080TI GPU上<4 ms）。提出的方法是可以解释的，因为它本地化了面部的哪一部分被标记为欺骗。实验结果表明，当在未知攻击下由13种不同的欺骗类型的数据集进行评估时，SSR-FCN可以实现TDR = 65％ @ 2.0％的FDR，同时在标准基准数据集（Oulu-NPU，Casia-MFSD，Casia-Mfsd和Repleay-Attack）下实现竞争性能。
State-of-the-art spoof detection methods tend to overfit to the spoof types seen during training and fail to generalize to unknown spoof types. Given that face anti-spoofing is inherently a local task, we propose a face anti-spoofing framework, namely Self-Supervised Regional Fully Convolutional Network (SSR-FCN), that is trained to learn local discriminative cues from a face image in a self-supervised manner. The proposed framework improves generalizability while maintaining the computational efficiency of holistic face anti-spoofing approaches (< 4 ms on a Nvidia GTX 1080Ti GPU). The proposed method is interpretable since it localizes which parts of the face are labeled as spoofs. Experimental results show that SSR-FCN can achieve TDR = 65% @ 2.0% FDR when evaluated on a dataset comprising of 13 different spoof types under unknown attacks while achieving competitive performances under standard benchmark datasets (Oulu-NPU, CASIA-MFSD, and Replay-Attack).
