LoRA-Enhanced Vision Transformer for Single Image based Morphing Attack Detection via Knowledge Distillation from EfficientNet

TL;DR

This paper introduces a LoRA-enhanced Vision Transformer model for single-image morphing attack detection, utilizing knowledge distillation from an EfficientNet-based teacher to improve accuracy and efficiency in face recognition security systems.

Contribution

It presents a novel ViT-based S-MAD method with LoRA fine-tuning and a teacher-student framework, achieving superior detection accuracy and computational efficiency over existing methods.

Findings

Outperforms six state-of-the-art S-MAD techniques in accuracy.

Demonstrates robustness across ten morphing algorithms.

Reduces computational costs with LoRA fine-tuning.

Abstract

Face Recognition Systems (FRS) are critical for security but remain vulnerable to morphing attacks, where synthetic images blend biometric features from multiple individuals. We propose a novel Single-Image Morphing Attack Detection (S-MAD) approach using a teacher-student framework, where a CNN-based teacher model refines a ViT-based student model. To improve efficiency, we integrate Low-Rank Adaptation (LoRA) for fine-tuning, reducing computational costs while maintaining high detection accuracy. Extensive experiments are conducted on a morphing dataset built from three publicly available face datasets, incorporating ten different morphing generation algorithms to assess robustness. The proposed method is benchmarked against six state-of-the-art S-MAD techniques, demonstrating superior detection performance and computational efficiency.