Evaluating the Effectiveness of Attack-Agnostic Features for Morphing Attack Detection

TL;DR

This paper investigates the use of attack-agnostic features from large vision models for detecting morphing attacks, demonstrating their effectiveness across various scenarios and outperforming traditional methods.

Contribution

It introduces a novel approach using pretrained vision model features with simple classifiers for morphing attack detection, showing improved robustness and generalization.

Findings

Attack-agnostic features outperform traditional detectors in most scenarios.

Supervised SVM and one-class GMM detectors effectively utilize these features.

The approach generalizes well to unseen attacks and different data sources.

Abstract

Morphing attacks have diversified significantly over the past years, with new methods based on generative adversarial networks (GANs) and diffusion models posing substantial threats to face recognition systems. Recent research has demonstrated the effectiveness of features extracted from large vision models pretrained on bonafide data only (attack-agnostic features) for detecting deep generative images. Building on this, we investigate the potential of these image representations for morphing attack detection (MAD). We develop supervised detectors by training a simple binary linear SVM on the extracted features and one-class detectors by modeling the distribution of bonafide features with a Gaussian Mixture Model (GMM). Our method is evaluated across a comprehensive set of attacks and various scenarios, including generalization to unseen attacks, different source datasets, and print-scan data. Our results indicate that attack-agnostic features can effectively detect morphing attacks, outperforming traditional supervised and one-class detectors from the literature in most scenarios. Additionally, we provide insights into the strengths and limitations of each considered representation and discuss potential future research directions to further enhance the robustness and generalizability of our approach.