XSA-MAD: Cross-modal Semantic Alignment for Morphing Attack Detection

• URL: http://arxiv.org/abs/2608.13861v1
• Date: Fri, 14 Aug 2026 01:16:43 GMT
• Title: XSA-MAD: Cross-modal Semantic Alignment for Morphing Attack Detection
• Abstract summary: XSA-MAD explicitly models semantic inconsistencies between bona-fide and morphed faces.<n>It consistently outperforms existing methods under high-fidelity generative attacks.
• Score: 5.108289909311743
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Morphing attacks pose a serious threat to face recognition systems. However, existing image-based morphing attack detection (MAD) methods often generalize poorly to unseen generation techniques because they rely solely on visual cues. We propose XSA-MAD, a CLIP-based multimodal framework that explicitly models semantic inconsistencies between bona-fide and morphed faces. Morphing concepts are decomposed into four interpretable attributes, including identity, facial geometry, texture, and consistency, and are encoded as structured and attribute-aware textual representations. The image encoder is progressively aligned with this discriminative textual space, resulting in a unified semantic representation that captures generation-invariant and concept-level discrepancies between bona-fide and morph images. Experiments on MAD22 and MorDIFF, following training on SMDD, demonstrate strong generalization across diverse morphing principles. In particular, XSA-MAD achieves an equal error rate of 2.92% on GAN-based morphs and consistently outperforms existing methods under high-fidelity generative attacks.

Related papers

• Face Re-morphing: Differential Morphing Attack Detection via Feature-Space Similarity Changes [5.986952206758139]
  Face morphing attacks pose a serious threat to face recognition systems.<n>This paper proposes Face Re-morphing, a D-MAD method that uses the feature-space response to an additional morphing operation as a detection cue.<n> Experiments on FRLL-Morphs and FEI Morph show that the proposed cue is effective across different morphing conditions.
  arXiv  Detail & Related papers  (2026-08-14T01:10:53Z)
• StableMorph: High-Quality Face Morph Generation with Stable Diffusion [9.45633874048024]
  Face morphing attacks threaten the integrity of biometric identity systems.<n>We introduce StableMorph, a novel approach that generates highly realistic, artifact-free morphed face images.<n>We show that StableMorph images not only rival or exceed the quality of genuine face images but also maintain a strong ability to fool face recognition systems.
  arXiv  Detail & Related papers  (2025-11-11T10:44:41Z)
• RAM++: Robust Representation Learning via Adaptive Mask for All-in-One Image Restoration [94.49712266736141]
  RAM++ is a two-stage framework for all-in-one image restoration.<n>It integrates high-level semantic understanding with low-level texture generation.<n>It addresses the limitations of existing degradation-oriented methods in extreme scenarios.
  arXiv  Detail & Related papers  (2025-09-15T15:24:15Z)
• Harmonizing Visual Representations for Unified Multimodal Understanding and Generation [53.01486796503091]
  We present emphHarmon, a unified autoregressive framework that harmonizes understanding and generation tasks with a shared MAR encoder.<n>Harmon achieves state-of-the-art image generation results on the GenEval, MJHQ30K and WISE benchmarks.
  arXiv  Detail & Related papers  (2025-03-27T20:50:38Z)
• Unsupervised Modality Adaptation with Text-to-Image Diffusion Models for Semantic Segmentation [54.96563068182733]
  We propose Modality Adaptation with text-to-image Diffusion Models (MADM) for semantic segmentation task. MADM utilizes text-to-image diffusion models pre-trained on extensive image-text pairs to enhance the model's cross-modality capabilities. We show that MADM achieves state-of-the-art adaptation performance across various modality tasks, including images to depth, infrared, and event modalities.
  arXiv  Detail & Related papers  (2024-10-29T03:49:40Z)
• LADIMO: Face Morph Generation through Biometric Template Inversion with Latent Diffusion [5.602947425285195]
  Face morphing attacks pose a severe security threat to face recognition systems. We present a representation-level face morphing approach, namely LADIMO, that performs morphing on two face recognition embeddings. We show that each face morph variant has an individual attack success rate, enabling us to maximize the morph attack potential.
  arXiv  Detail & Related papers  (2024-10-10T14:41:37Z)
• Face Morphing Attack Detection with Denoising Diffusion Probabilistic Models [0.0]
  Morphed face images can be used to impersonate someone's identity for various malicious purposes. Existing MAD techniques rely on discriminative models that learn from examples of bona fide and morphed images. We propose a novel, diffusion-based MAD method that learns only from the characteristics of bona fide images.
  arXiv  Detail & Related papers  (2023-06-27T18:19:45Z)
• MorphGANFormer: Transformer-based Face Morphing and De-Morphing [55.211984079735196]
  StyleGAN-based approaches to face morphing are among the leading techniques. We propose a transformer-based alternative to face morphing and demonstrate its superiority to StyleGAN-based methods.
  arXiv  Detail & Related papers  (2023-02-18T19:09:11Z)
• MorDIFF: Recognition Vulnerability and Attack Detectability of Face Morphing Attacks Created by Diffusion Autoencoders [10.663919597506055]
  Face morphing attacks are created on the image-level or on the representation-level. Recent advances in the diffusion autoencoder models have overcome the GAN limitations, leading to high reconstruction fidelity. This work investigates using diffusion autoencoders to create face morphing attacks by comparing them to a wide range of image-level and representation-level morphs.
  arXiv  Detail & Related papers  (2023-02-03T16:37:38Z)
• Siamese Image Modeling for Self-Supervised Vision Representation Learning [73.78790119050056]
  Self-supervised learning (SSL) has delivered superior performance on a variety of downstream vision tasks. Two main-stream SSL frameworks have been proposed, i.e., Instance Discrimination (ID) and Masked Image Modeling (MIM) We propose Siamese Image Modeling (SIM), which predicts the dense representations of an augmented view.
  arXiv  Detail & Related papers  (2022-06-02T17:59:58Z)
• Multi-Margin based Decorrelation Learning for Heterogeneous Face Recognition [90.26023388850771]
  This paper presents a deep neural network approach to extract decorrelation representations in a hyperspherical space for cross-domain face images. The proposed framework can be divided into two components: heterogeneous representation network and decorrelation representation learning. Experimental results on two challenging heterogeneous face databases show that our approach achieves superior performance on both verification and recognition tasks.
  arXiv  Detail & Related papers  (2020-05-25T07:01:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.