SFDemorpher: Generalizable Face Demorphing for Operational Morphing Attack Detection

2026-03-30

AI Summary

This work addresses the critical security threat posed by face morphing attacks, which generate verifiable identity documents containing multiple identities and thereby compromise biometric systems. Existing detection methods suffer from limited generalization in real-world scenarios. To overcome this, the paper proposes SFDemorpher, a novel framework that jointly disentangles identity information in both the StyleGAN latent space and high-dimensional feature space. Leveraging a two-stage training strategy and a synthetic-to-real hybrid dataset, SFDemorpher achieves strong generalization across unseen identities, diverse capture conditions, and 13 distinct morphing techniques—the first method to do so. It substantially widens the score distribution gap between genuine and morphed samples, attaining state-of-the-art performance in practical applications such as border control and identity document enrollment, while also enabling high-fidelity visual explanations.

Abstract

Face morphing attacks compromise biometric security by creating document images that verify against multiple identities, posing significant risks from document issuance to border control. Differential Morphing Attack Detection (D-MAD) offers an effective countermeasure, particularly when employing face demorphing to disentangle identities blended in the morph. However, existing methods lack operational generalizability due to limited training data and the assumption that all document inputs are morphs. This paper presents SFDemorpher, a framework designed for the operational deployment of face demorphing for D-MAD that performs identity disentanglement within joint StyleGAN latent and high-dimensional feature spaces. We introduce a dual-pass training strategy handling both morphed and bona fide documents, leveraging a hybrid corpus with predominantly synthetic identities to enhance robustness against unseen distributions. Extensive evaluation confirms state-of-the-art generalizability across unseen identities, diverse capture conditions, and 13 morphing techniques, spanning both border verification and the challenging document enrollment stage. Our framework achieves superior D-MAD performance by widening the margin between the score distributions of bona fide and morphed samples while providing high-fidelity visual reconstructions facilitating explainability.

Problem

• face demorphing
• morphing attack detection
• generalizability
• biometric security
• operational deployment

Innovation

• face demorphing
• StyleGAN latent space
• differential morphing attack detection
• generalizable biometric security
• synthetic identity training