Learning Multiple Explainable and Generalizable Cues for Face Anti-spoofing

TL;DR

This paper introduces a novel framework called MEGC that learns multiple explainable cues such as spoof boundary, moiré pattern, reflection artifacts, and facial depth to improve the generalization of face anti-spoofing models across datasets.

Contribution

The paper proposes a new multi-cue learning framework for face anti-spoofing that incorporates synthetic auxiliary supervision maps inspired by human decision cues.

Findings

Achieves state-of-the-art performance on public datasets.

Demonstrates improved cross-dataset generalization.

Validates effectiveness of multiple auxiliary cues.

Abstract

Although previous CNN based face anti-spoofing methods have achieved promising performance under intra-dataset testing, they suffer from poor generalization under cross-dataset testing. The main reason is that they learn the network with only binary supervision, which may learn arbitrary cues overfitting on the training dataset. To make the learned feature explainable and more generalizable, some researchers introduce facial depth and reflection map as the auxiliary supervision. However, many other generalizable cues are unexplored for face anti-spoofing, which limits their performance under cross-dataset testing. To this end, we propose a novel framework to learn multiple explainable and generalizable cues (MEGC) for face anti-spoofing. Specifically, inspired by the process of human decision, four mainly used cues by humans are introduced as auxiliary supervision including the boundary of spoof medium, moir\'e pattern, reflection artifacts and facial depth in addition to the binary supervision. To avoid extra labelling cost, corresponding synthetic methods are proposed to generate these auxiliary supervision maps. Extensive experiments on public datasets validate the effectiveness of these cues, and state-of-the-art performances are achieved by our proposed method.