Arc2Morph: Identity-Preserving Facial Morphing with Arc2Face

TL;DR

Experimental results show that the proposed deep learning-based approach achieves a morphing attack potential comparable to that of landmark-based techniques, which have traditionally been regarded as the most challenging.

Abstract

Face morphing attacks are widely recognized as one of the most challenging threats to face recognition systems used in electronic identity documents. These attacks exploit a critical vulnerability in passport enrollment procedures adopted by many countries, where the facial image is often acquired without a supervised live capture process. In this paper, we propose a novel face morphing technique based on Arc2Face, an identity-conditioned face foundation model capable of synthesizing photorealistic facial images from compact identity representations. We demonstrate the effectiveness of the proposed approach by comparing the morphing attack potential metric on two large-scale sequestered face morphing attack detection datasets against several state-of-the-art morphing methods, as well as on two novel morphed face datasets derived from FEI and ONOT. Experimental results show that the proposed deep learning-based approach achieves a morphing attack potential comparable to that of landmark-based techniques, which have traditionally been regarded as the most challenging. These findings confirm the ability of the proposed method to effectively preserve and manage identity information during the morph generation process.

Figures (6)

• Figure 1: Samples of morphed images (center column) generated with our proposed method. On the first row, Accomplice (left column) and criminal (right column) are sourced from the ONOT di2024onot dataset; on the second row, accomplice and criminal are sourced from the FEI Fei_THOMAZ2010902 dataset.
• Figure 2: Overview of the proposed method. The input images $I_A$ and $I_B$ are first encoded using the identity encoder $E_id(\cdot)$, producing the identity embeddings $e_A$ and $e_B$. In parallel, we extract the pose conditioning image for Arc2Face with EMOCAv2 danvevcek2022emocafilntisis2022visual, obtaining $I_P$. Then, the two compact identity representations $e_A$ and $e_B$ are mapped into the CLIP latent space and interpolated to obtain the latent representation $p_M$, which is then decoded alongside with the conditioning image $I_P$ to generate the morphed image $I_M'$. Finally, the resulting image is post-processed using BEN2 to remove the background, yielding the final morphed image $I_M$.
• Figure 3: Visualization of the MAPs computed on the MONOT dataset for the proposed approach and the competitors, considering ten "in the wild" images as probe.
• Figure 4: Visualization of the MAPs computed on the EINMorph-HQ v2 dataset for the proposed approach and the competitors.
• Figure 5: Visualization of the MAPs computed on the EINMorph-MQ v2 dataset for the proposed approach and the competitors.