On the detection of morphing attacks generated by GANs

TL;DR

This paper investigates the detection of GAN-generated face morphs, proposing simple spectral and LBP-based methods, and finds that a pretrained ResNet combined with LBP features achieves near-perfect accuracy, highlighting the importance of diverse detection strategies.

Contribution

It introduces and evaluates simple spectral, LBP, and CNN-based detectors for GAN morphs, demonstrating the effectiveness of LBP+ResNet fusion for robust detection.

Findings

LBP-based systems perform well intra-dataset but less so cross-dataset.

Fusion of LBP and CNN improves detection accuracy.

Pretrained ResNet achieves near-perfect detection of GAN morphs.

Abstract

Recent works have demonstrated the feasibility of GAN-based morphing attacks that reach similar success rates as more traditional landmark-based methods. This new type of "deep" morphs might require the development of new adequate detectors to protect face recognition systems. We explore simple deep morph detection baselines based on spectral features and LBP histograms features, as well as on CNN models, both in the intra-dataset and cross-dataset case. We observe that simple LBP-based systems are already quite accurate in the intra-dataset setting, but struggle with generalization, a phenomenon that is partially mitigated by fusing together several of those systems at score-level. We conclude that a pretrained ResNet effective for GAN image detection is the most effective overall, reaching close to perfect accuracy. We note however that LBP-based systems maintain a level of interest : additionally to their lower computational requirements and increased interpretability with respect to CNNs, LBP+ResNet fusions sometimes also showcase increased performance versus ResNet-only, hinting that LBP-based systems can focus on meaningful signal that is not necessarily picked up by the CNN detector.