Deep Anomaly Detection for Generalized Face Anti-Spoofing
This addresses the generalization challenge in face anti-spoofing for real-world security applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of generalized face anti-spoofing by reformulating it as an anomaly detection task, achieving state-of-the-art results with a deep metric learning model that outperforms existing methods by a considerable margin on the GRAD-GPAD dataset.
Face recognition has achieved unprecedented results, surpassing human capabilities in certain scenarios. However, these automatic solutions are not ready for production because they can be easily fooled by simple identity impersonation attacks. And although much effort has been devoted to develop face anti-spoofing models, their generalization capacity still remains a challenge in real scenarios. In this paper, we introduce a novel approach that reformulates the Generalized Presentation Attack Detection (GPAD) problem from an anomaly detection perspective. Technically, a deep metric learning model is proposed, where a triplet focal loss is used as a regularization for a novel loss coined "metric-softmax", which is in charge of guiding the learning process towards more discriminative feature representations in an embedding space. Finally, we demonstrate the benefits of our deep anomaly detection architecture, by introducing a few-shot a posteriori probability estimation that does not need any classifier to be trained on the learned features. We conduct extensive experiments using the GRAD-GPAD framework that provides the largest aggregated dataset for face GPAD. Results confirm that our approach is able to outperform all the state-of-the-art methods by a considerable margin.