ML Security PapersR-FLoRA: Residual-Statistic-Gated Low-Rank Adaptation for Single-Image Face Morphing Attack Detection
Published on arXiv
2604.17321
Input Manipulation Attack
OWASP ML Top 10 — ML01
Key Finding
Consistently surpasses nine recent state-of-the-art S-MAD algorithms in detection accuracy and cross-domain generalization with real-time efficiency using frozen backbone and minimal trainable parameters
R-FLoRA
Novel technique introduced
Face morphing attacks pose a substantial risk to the reliability of face recognition systems used in passport issuance, border control, and digital identity verification. Detecting morphing attacks from a single facial image remains challenging owing to the lack of a trusted reference and the diversity of attack generation methods. This paper presents a new Single-Image Face Morphing Attack Detection (S-MAD) framework that integrates high-frequency Laplacian residual statistics with representations from a frozen, foundation-scale vision transformer. The approach employs residual-statistic-gated low-rank adapters (R-FLoRA) and feature-wise residual fusion (Res-FiLM) to enhance sensitivity to local morphing artefacts while preserving the semantic context of the backbone. A novel residual-contrastive alignment loss further regularises the fused token space, improving discrimination under unseen morphing conditions. Comprehensive experiments on four ICAO-compliant datasets, encompassing seven morph generation techniques, demonstrate that the proposed method consistently surpasses nine recent state-of-the-art S-MAD algorithms in detection accuracy and cross-domain (or dataset) generalisation. With a frozen backbone and minimal trainable parameters, the model achieves real-time efficiency and interpretability, making it suitable for real-life scenarios in biometric verification systems.
Key Contributions
- Residual-statistic-gated low-rank adapters (R-FLoRA) that integrate high-frequency Laplacian residual statistics with frozen vision transformer representations
- Feature-wise residual fusion (Res-FiLM) and residual-contrastive alignment loss for enhanced morphing artifact detection
- Cross-dataset generalization across four ICAO-compliant datasets with seven morph generation techniques, outperforming nine state-of-the-art S-MAD methods
🛡️ Threat Analysis
Face morphing attacks are adversarial manipulations of input images designed to evade face recognition systems at inference time. The paper proposes a defense (detection method) against these input manipulation attacks on biometric verification systems.