Face Morphing Attack Detection Using Privacy-Aware Training Data

TL;DR

This paper investigates face morphing attack detection using privacy-preserving synthetic training data, demonstrating that models trained solely on non-existing people's faces can effectively detect real-world morphing attacks.

Contribution

It introduces a novel approach of training detection algorithms exclusively on synthetic non-existing faces, addressing privacy concerns and data scarcity.

Findings

Synthetic training data generalizes well to real-world datasets

Detection algorithms trained on synthetic data achieve high accuracy

Privacy-preserving training is effective for morphing attack detection

Abstract

Images of morphed faces pose a serious threat to face recognition--based security systems, as they can be used to illegally verify the identity of multiple people with a single morphed image. Modern detection algorithms learn to identify such morphing attacks using authentic images of real individuals. This approach raises various privacy concerns and limits the amount of publicly available training data. In this paper, we explore the efficacy of detection algorithms that are trained only on faces of non--existing people and their respective morphs. To this end, two dedicated algorithms are trained with synthetic data and then evaluated on three real-world datasets, i.e.: FRLL-Morphs, FERET-Morphs and FRGC-Morphs. Our results show that synthetic facial images can be successfully employed for the training process of the detection algorithms and generalize well to real-world scenarios.