mlsanityData to 7 Oct 2026

MA-ViT: Modality-Agnostic Vision Transformers for Face Anti-Spoofing

arXiv 2304.07549 ↗First posted 15 Apr 2023cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

The existing multi-modal face anti-spoofing (FAS) frameworks are designed based on two strategies: halfway and late fusion. However, the former requires test modalities consistent with the training input, which seriously limits its deployment scenarios. And the latter is built on multiple branches to process different modalities independently, which limits their use in applications with low memory or fast execution requirements. In this work, we present a single branch based Transformer framework, namely Modality-Agnostic Vision Transformer (MA-ViT), which aims to improve the performance of arbitrary modal attacks with the help of multi-modal data. Specifically, MA-ViT adopts the early fusion to aggregate all the available training modalities data and enables flexible testing of any given modal samples. Further, we develop the Modality-Agnostic Transformer Block (MATB) in MA-ViT, which consists of two stacked attentions named Modal-Disentangle Attention (MDA) and Cross-Modal Attention (CMA), to eliminate modality-related information for each modal sequences and supplement modality-agnostic liveness features from another modal sequences, respectively. Experiments demonstrate that the single model trained based on MA-ViT can not only flexibly evaluate different modal samples, but also outperforms existing single-modal frameworks by a large margin, and approaches the multi-modal frameworks introduced with smaller FLOPs and model parameters.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 19 of its 25 references that are in the map. First posted on arXiv 15 Apr 2023.

Builds on

19 of 25 references.

References · 19

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  3. 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
  4. 2020Searching Central Difference Convolutional Networks for Face Anti-Spoofingcited by 173
  5. 2018Learning Deep Models for Face Anti-Spoofing: Binary or Auxiliary Supervisioncited by 167
  6. 2021CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptationcited by 129
  7. 2019Biometric Face Presentation Attack Detection with Multi-Channel Convolutional Neural Networkcited by 78
  8. 2019Deep Pixel-wise Binary Supervision for Face Presentation Attack Detectioncited by 76
  9. 2020CASIA-SURF CeFA: A Benchmark for Multi-modal Cross-ethnicity Face Anti-spoofingcited by 66
  10. 2020Face Anti-Spoofing Via Disentangled Representation Learningcited by 65
  11. 2021KVT: k-NN Attention for Boosting Vision Transformerscited by 64
  12. 2020On Disentangling Spoof Trace for Generic Face Anti-Spoofingcited by 62
7 more
  1. 2019CASIA-SURF: A Large-scale Multi-modal Benchmark for Face Anti-spoofingcited by 61
  2. 2018A Dataset and Benchmark for Large-scale Multi-modal Face Anti-spoofingcited by 51
  3. 2021Contrastive Context-Aware Learning for 3D High-Fidelity Mask Face Presentation Attack Detectioncited by 43
  4. 2021Cross Modal Focal Loss for RGBD Face Anti-Spoofingcited by 41
  5. 2020Learning One Class Representations for Face Presentation Attack Detection using Multi-channel Convolutional Neural Networkscited by 36
  6. 2020Cross-ethnicity Face Anti-spoofing Recognition Challenge: A Reviewcited by 25
  7. 20213D High-Fidelity Mask Face Presentation Attack Detection Challengecited by 18

Built on by

Cited by year

Citing papers per year: 2023: 6; 2024: 8; 2025: 4.

18 papers.

Sort:

By area:

  1. 2023Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detectioncited by 180
  2. 2024SHIELD : An Evaluation Benchmark for Face Spoofing and Forgery Detection with Multimodal Large Language Modelscited by 39
  3. 2024CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofingcited by 39
  4. 2023FM-ViT: Flexible Modal Vision Transformers for Face Anti-Spoofingcited by 22
  5. 2024Test-Time Domain Generalization for Face Anti-Spoofingcited by 20
  6. 2024Unified Physical-Digital Face Attack Detectioncited by 17
  7. 2023Unified Frequency-Assisted Transformer Framework for Detecting and Grounding Multi-Modal Manipulationcited by 14
  8. 2023Visual Prompt Flexible-Modal Face Anti-Spoofingcited by 9
  9. 2024Joint Physical-Digital Facial Attack Detection Via Simulating Spoofing Cluescited by 8
  10. 2024La-SoftMoE CLIP for Unified Physical-Digital Face Attack Detectioncited by 8
  11. 2025Domain Generalization for Face Anti-spoofing via Content-aware Composite Prompt Engineeringcited by 7
  12. 2025DADM: Dual Alignment of Domain and Modality for Face Anti-spoofingcited by 4
6 more
  1. 2024Unified Physical-Digital Attack Detection Challengecited by 3
  2. 2025FA^{3}-CLIP: Frequency-Aware Cues Fusion and Attack-Agnostic Prompt Learning for Unified Face Attack Detectioncited by 3
  3. 2025SUEDE:Shared Unified Experts for Physical-Digital Face Attack Detection Enhancementcited by 2
  4. 2023Hyperbolic Face Anti-Spoofingcited by 1
  5. 2024BIG-MoE: Bypass Isolated Gating MoE for Generalized Multimodal Face Anti-Spoofingcited by 1
  6. 2023CLIPC8: Face liveness detection algorithm based on image-text pairs and contrastive learningcited by 0