Generalized Face Liveness Detection via De-fake Face Generator

TL;DR

This paper introduces a novel face liveness detection method leveraging a large-scale real face dataset and a de-fake face generator to improve generalization to unseen spoofing attacks and domains.

Contribution

It proposes the AG-FAS framework with a de-fake face generator and an off-real attention network, enhancing fake feature learning and cross-domain robustness.

Findings

Achieves state-of-the-art results on nine public datasets.

Effectively generalizes to unseen domains and attack types.

Provides theoretical analysis of anomalous cues.

Abstract

Previous Face Anti-spoofing (FAS) methods face the challenge of generalizing to unseen domains, mainly because most existing FAS datasets are relatively small and lack data diversity. Thanks to the development of face recognition in the past decade, numerous real face images are available publicly, which are however neglected previously by the existing literature. In this paper, we propose an Anomalous cue Guided FAS (AG-FAS) method, which can effectively leverage large-scale additional real faces for improving model generalization via a De-fake Face Generator (DFG). Specifically, by training on a large-scale real face only dataset, the generator obtains the knowledge of what a real face should be like, and thus has the capability of generating a "real" version of any input face image. Consequently, the difference between the input face and the generated "real" face can be treated as cues of attention for the fake feature learning. With the above ideas, an Off-real Attention Network (OA-Net) is proposed which allocates its attention to the spoof region of the input according to the anomalous cue. Extensive experiments on a total of nine public datasets show our method achieves state-of-the-art results under cross-domain evaluations with unseen scenarios and unknown presentation attacks. Besides, we provide theoretical analysis demonstrating the effectiveness of the proposed anomalous cues.