A Survey of Domain Generalization-Based Face Anti-spoofing

Fangling Jiang, Yunfan Liu, Bing Liu, Xiaoliang Chen, Qi Li

Abstract: In recent years, remarkable research attention has been attracted to improve the generalization ability of face anti-spoofing methods, and domain generalization techniques have been widely exploited for adapting face anti-spoofing models to unseen testing scenarios. In this paper, we present a comprehensive survey on domain generalization-based face anti-spoofing methods. Specifically, we propose a taxonomy for existing methods and conduct a thorough review on these methods by comparing and analyzing their motivations, highlights, and common technical characteristics. Afterward, we introduce commonly used datasets and evaluation metrics, and also analyze the performance of existing methods to uncover key factors affecting the generalization performance. Finally, we conclude this survey with a forecast on promising future research directions.