Deep Learning for Face Anti-Spoofing: A Survey

Abstract

Face anti-spoofing (FAS) has lately attracted increasing attention due to its vital role in securing face recognition systems from presentation attacks (PAs). As more and more realistic PAs with novel types spring up, early-stage FAS methods based on handcrafted features become unreliable due to their limited representation capacity. With the emergence of large-scale academic datasets in the recent decade, deep learning based FAS achieves remarkable performance and dominates this area. However, existing reviews in this field mainly focus on the handcrafted features, which are outdated and uninspiring for the progress of FAS community. In this paper, to stimulate future research, we present the first comprehensive review of recent advances in deep learning based FAS. It covers several novel and insightful components: 1) besides supervision with binary label (e.g., '0' for bonafide vs. '1' for PAs), we also investigate recent methods with pixel-wise supervision (e.g., pseudo depth map); 2) in addition to traditional intra-dataset evaluation, we collect and analyze the latest methods specially designed for domain generalization and open-set FAS; and 3) besides commercial RGB camera, we summarize the deep learning applications under multi-modal (e.g., depth and infrared) or specialized (e.g., light field and flash) sensors. We conclude this survey by emphasizing current open issues and highlighting potential prospects.

Learning Meta Pattern for Face Anti-Spoofing

“…Due to the powerful representation ability of deep neural networks, deep learning based FAS methods gradually surpass and replace the traditional methods based on handcrafted features [39,49,19]. Recently, plenty of domain generalization and adaptation techniques have been proposed to improve the cross-domain FAS performance.…”

“…Face recognition (FR) has been widely used in identity authentication because of its convenience. However, face recognition-based authentication systems are threatened by face spoofing attacks [39,19,49]. To protect FR systems from spoofing attacks, Face Anti-Spoofing (FAS) techniques are deployed to detect spoofing faces and reject malicious attempts.…”

Comparative Analysis of Vulnerabilities in Biometric Face Recognition Systems

“…Иногда спроектированные пространства признаков не в состоянии отличить подделку от артефактов лица. Именно поэтому внимание сообщества исследователей привлекли нейронные сети, в частности свёрточные сети и сети глубокого обучения [49].…”

Face spoofing detection model based on multi-scale predictive feature fusion

“…However, the vulnerability to some attacks such as printed photos, video replay, and 3D masks limits its reliable deployment. Face Anti-Spoofing (FAS) [1,2] ，also known as face spoofing detection method, is used to distinguish between real and fake faces and ensure that only real faces are matched, which is extremely important for the security of face recognition systems.…”