Generalized Disguise Makeup Presentation Attack Detection Using an Attention-Guided Patch-Based Framework

TL;DR

This paper introduces a novel attention-guided patch-based framework for detecting disguise makeup presentation attacks in facial recognition, utilizing a new diverse dataset and achieving state-of-the-art results.

Contribution

It presents a generalized, style-invariant detection method with a two-phase approach and introduces a new dataset for disguise makeup attack detection.

Findings

Achieved 8.97% ACER and 9.76% EER on their dataset.

Achieved 0% ACER on Obfuscation and Impersonation attacks in SIW-Mv2.

Outperformed prior methods in disguise makeup attack detection.

Abstract

Despite significant advances in facial recognition systems, they remain vulnerable to face presentation attacks. Among them, disguise makeup attacks are particularly challenging, as they use advanced cosmetics, prosthetic components, and artificial materials to realistically alter facial appearance, often making detection difficult even for humans. Despite their importance, this problem remains underexplored, and publicly available datasets are limited. To address this, we propose a generalized disguise makeup presentation attack detection framework. The method adopts a two-phase design in which a style-invariant full-face model, trained with metric learning and enhanced by a whitening transformation, extracts region attention scores via Grad-CAM. These scores guide a patch-based phase that performs localized analysis using region-specific subnetworks trained with metric learning for fine-grained discrimination. We also construct a new, diverse dataset of live and disguise makeup faces collected under real-world conditions, covering variations in subjects, environments, and disguise materials. Experimental results demonstrate strong generalization across both the collected dataset and SIW-Mv2, achieving 8.97% ACER and 9.76% EER on the collected dataset, and 0% ACER on Obfuscation and Impersonation and 1.34% on Cosmetics attacks of SIW-Mv2. The proposed method consistently outperforms prior works while maintaining robust performance across other spoof types.