Domain-generalizable Face Anti-Spoofing with Patch-based Multi-tasking and Artifact Pattern Conversion
2026-04-10 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address problems with face anti-spoofing systems, which have trouble recognizing fake faces in new situations due to limited training data. They created PCGAN, a system that helps generate fake face images with different patterns to train the model better. This system separates features related to fake artifacts and real faces, improving detection of partial face attacks. Their tests show PCGAN can better handle unseen spoofing methods and improve face recognition security.
Face Anti-SpoofingGenerative Adversarial NetworkDomain GeneralizationLatent VectorsPatch-based LearningMulti-task LearningSpoof ArtifactsPartial AttacksFace Recognition Security
Authors
Seungjin Jung, Yonghyun Jeong, Minha Kim, Jimin Min, Youngjoon Yoo, Jongwon Choi
Abstract
Face Anti-Spoofing (FAS) algorithms, designed to secure face recognition systems against spoofing, struggle with limited dataset diversity, impairing their ability to handle unseen visual domains and spoofing methods. We introduce the Pattern Conversion Generative Adversarial Network (PCGAN) to enhance domain generalization in FAS. PCGAN effectively disentangles latent vectors for spoof artifacts and facial features, allowing to generate images with diverse artifacts. We further incorporate patch-based and multi-task learning to tackle partial attacks and overfitting issues to facial features. Our extensive experiments validate PCGAN's effectiveness in domain generalization and detecting partial attacks, giving a substantial improvement in facial recognition security.