Instance-Aware Domain Generalization for Face Anti-Spoofing

TL;DR

This paper introduces an instance-aware domain generalization framework for face anti-spoofing that aligns features at the instance level to improve generalization across unseen scenarios, without relying on coarse domain labels.

Contribution

It proposes a novel instance-aware approach with asymmetric whitening and style diversification techniques to learn style-insensitive features for face anti-spoofing.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effectively reduces sensitivity to style variations.

Enhances generalization to unseen domains.

Abstract

Face anti-spoofing (FAS) based on domain generalization (DG) has been recently studied to improve the generalization on unseen scenarios. Previous methods typically rely on domain labels to align the distribution of each domain for learning domain-invariant representations. However, artificial domain labels are coarse-grained and subjective, which cannot reflect real domain distributions accurately. Besides, such domain-aware methods focus on domain-level alignment, which is not fine-grained enough to ensure that learned representations are insensitive to domain styles. To address these issues, we propose a novel perspective for DG FAS that aligns features on the instance level without the need for domain labels. Specifically, Instance-Aware Domain Generalization framework is proposed to learn the generalizable feature by weakening the features' sensitivity to instance-specific styles. Concretely, we propose Asymmetric Instance Adaptive Whitening to adaptively eliminate the style-sensitive feature correlation, boosting the generalization. Moreover, Dynamic Kernel Generator and Categorical Style Assembly are proposed to first extract the instance-specific features and then generate the style-diversified features with large style shifts, respectively, further facilitating the learning of style-insensitive features. Extensive experiments and analysis demonstrate the superiority of our method over state-of-the-art competitors. Code will be publicly available at https://github.com/qianyuzqy/IADG.