mlsanityData to 7 Oct 2026

Cross Modal Focal Loss for RGBD Face Anti-Spoofing

arXiv 2103.00948 ↗First posted 1 Mar 2021cs.CV cs.CR cs.LGcited by 41

Not on the map. Why?

Abstract (arXiv)

Automatic methods for detecting presentation attacks are essential to ensure the reliable use of facial recognition technology. Most of the methods available in the literature for presentation attack detection (PAD) fails in generalizing to unseen attacks. In recent years, multi-channel methods have been proposed to improve the robustness of PAD systems. Often, only a limited amount of data is available for additional channels, which limits the effectiveness of these methods. In this work, we present a new framework for PAD that uses RGB and depth channels together with a novel loss function. The new architecture uses complementary information from the two modalities while reducing the impact of overfitting. Essentially, a cross-modal focal loss function is proposed to modulate the loss contribution of each channel as a function of the confidence of individual channels. Extensive evaluations in two publicly available datasets demonstrate the effectiveness of the proposed approach.

Key facts. Cited by 41 papers in the map (data through 7 Oct 2026). Builds on 12 of its 31 references that are in the map. First posted on arXiv 1 Mar 2021.

Builds on

12 of 31 references.

References · 12

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016Densely Connected Convolutional Networkscited by 6,691
  3. 2016Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networkscited by 895
  4. 2015face anti-spoofing based on color texture analysiscited by 82
  5. 2019Biometric Face Presentation Attack Detection with Multi-Channel Convolutional Neural Networkcited by 78
  6. 2019Deep Pixel-wise Binary Supervision for Face Presentation Attack Detectioncited by 76
  7. 2019Deep Tree Learning for Zero-shot Face Anti-Spoofingcited by 73
  8. 2020Learning One Class Representations for Face Presentation Attack Detection using Multi-channel Convolutional Neural Networkscited by 36
  9. 2020Deep Models and Shortwave Infrared Information to Detect Face Presentation Attackscited by 29
  10. 2019Domain Adaptation in Multi-Channel Autoencoder based Features for Robust Face Anti-Spoofingcited by 12
  11. 2020The High-Quality Wide Multi-Channel Attack (HQ-WMCA) databasecited by 9
  12. 2020Can Your Face Detector Do Anti-spoofing? Face Presentation Attack Detection with a Multi-Channel Face Detectorcited by 7

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 9; 2022: 6; 2023: 12; 2024: 3; 2025: 7; 2026: 4.

41 papers.

Sort:

By area:

  1. 2021Deep Learning for Face Anti-Spoofing: A Surveycited by 64
  2. 2023DFormer: Rethinking RGBD Representation Learning for Semantic Segmentationcited by 58
  3. 2021Contrastive Context-Aware Learning for 3D High-Fidelity Mask Face Presentation Attack Detectioncited by 43
  4. 2024CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofingcited by 39
  5. 2023FM-ViT: Flexible Modal Vision Transformers for Face Anti-Spoofingcited by 22
  6. 2022Flexible-Modal Face Anti-Spoofing: A Benchmarkcited by 20
  7. 2023Rethinking Vision Transformer and Masked Autoencoder in Multimodal Face Anti-Spoofingcited by 19
  8. 2023MA-ViT: Modality-Agnostic Vision Transformers for Face Anti-Spoofingcited by 18
  9. 2022Digital and Physical Face Attacks: Reviewing and One Step Furthercited by 17
  10. 2021Exploring Depth Contribution for Camouflaged Object Detectioncited by 14
  11. 2021Consistency Regularization for Deep Face Anti-Spoofingcited by 10
  12. 2023Visual Prompt Flexible-Modal Face Anti-Spoofingcited by 9
29 more
  1. 2021FRT-PAD: Effective Presentation Attack Detection Driven by Face Related Taskcited by 7
  2. 2022A Comprehensive Evaluation on Multi-channel Biometric Face Presentation Attack Detectioncited by 7
  3. 2025Domain Generalization for Face Anti-spoofing via Content-aware Composite Prompt Engineeringcited by 7
  4. 2021Which is Making the Contribution: Modulating Unimodal and Cross-modal Dynamics for Multimodal Sentiment Analysiscited by 5
  5. 2021MutualFormer: Multi-Modality Representation Learning via Cross-Diffusion Attentioncited by 5
  6. 2021High-Accuracy RGB-D Face Recognition via Segmentation-Aware Face Depth Estimation and Mask-Guided Attention Networkcited by 5
  7. 2023M3FAS: An Accurate and Robust MultiModal Mobile Face Anti-Spoofing Systemcited by 5
  8. 2024Confidence Aware Learning for Reliable Face Anti-spoofingcited by 5
  9. 2025Mixture-of-Attack-Experts with Class Regularization for Unified Physical-Digital Face Attack Detectioncited by 5
  10. 2023Fine-Grained Annotation for Face Anti-Spoofingcited by 4
  11. 2024Generalized Face Liveness Detection via De-fake Face Generatorcited by 4
  12. 2025DADM: Dual Alignment of Domain and Modality for Face Anti-spoofingcited by 4
  13. 2022Learning Polysemantic Spoof Trace: A Multi-Modal Disentanglement Network for Face Anti-spoofingcited by 3
  14. 2023Liveness score-based regression neural networks for face anti-spoofingcited by 3
  15. 2023AdvFAS: A robust face anti-spoofing framework against adversarial examplescited by 3
  16. 2025FA^{3}-CLIP: Frequency-Aware Cues Fusion and Attack-Agnostic Prompt Learning for Unified Face Attack Detectioncited by 3
  17. 2022Attacking Face Recognition with T-shirts: Database, Vulnerability Assessment and Detectioncited by 2
  18. 2025Towards Robust Optical-SAR Object Detection under Missing Modalities: A Dynamic Quality-Aware Fusion Frameworkcited by 2
  19. 2022A Survey of Face Recognitioncited by 1
  20. 2023Flexible-modal Deception Detection with Audio-Visual Adaptercited by 1
  21. 2023Hyperbolic Face Anti-Spoofingcited by 1
  22. 2023Multi-Channel Cross Modal Detection of Synthetic Face Imagescited by 1
  23. 2021MToFNet: Object Anti-Spoofing with Mobile Time-of-Flight Datacited by 0
  24. 2025Multi-Modal Face Anti-Spoofing via Cross-Modal Feature Transitionscited by 0
  25. 2025Robust Multi-Modal Face Anti-Spoofing with Domain Adaptation: Tackling Missing Modalities, Noisy Pseudo-Labels, and Model Degradationcited by 0
  26. 2026Dynamic Graph Neural Network with Adaptive Features Selection for RGB-D Based Indoor Scene Recognitioncited by 0
  27. 2026DifFoundMAD: Foundation Models meet Differential Morphing Attack Detectioncited by 0
  28. 2026Detection of T-shirt Presentation Attacks in Face Recognition Systemscited by 0
  29. 2026GBU-Palm: A Multimodal Video Dataset and Benchmark for Palm Presentation Attack Detectioncited by 0