PoolAtnRes: Towards Generalisable Differential Morphing Attack Detection

Morphing attacks can successfully deceive face recognition systems, resulting in unreliable access control, especially in the border control scenario. Consequently, the development of Morphing Attack Detection (MAD) algorithms is crucial for detecting morphing attacks based on either a single facial image (S-MAD) or two facial images (Differential-MAD or D-MAD). In this work, we proposed a novel D-MAD approach, PoolAtnRes, to reliably detect morphing attacks. The proposed PoolAtnRes architecture is constructed using three main functional blocks, namely convolution pooling, Hybrid Attention and Residual blocks, which are serially connected to detect morphing attacks. Extensive experiments were performed on the newly constructed morphing dataset using nine morphing-generation techniques. The detection performance of the proposed PoolAtnRes model was compared with three state-of-the-art (SOTA) D-MAD techniques with different performance evaluation protocols to benchmark its generalizabllity to unseen morphing generation. The results obtained indicated the best performance of the proposed PoolAtnRes D-MAD.

1. Introduction

Face Recognition Systems (FRS) have consistently ex-hibited vulnerability to morphing attacks, which under-mines the reliability of face biometric systems widely utilized in secure access control applications. NIST Face Analysis Technology Evaluation (FATE) Morph [16] indi-cates that the higher the accuracy of FRS, the lesser the resistance to morphing attacks. Morphing is the process of blending multiple facial images to generate a compos-ite image that displays the texture and geometric features of source images. Given its complexity, it is difficult to distin-guish morphed images from skilled human observers, such as border guards [11], as well as automatic detection sys-tems [16].