Awesome-FAS (Face Authentication Security)
A curated list of Face Authentication Security (including face anti-spoofing/face presentation attack/face liveness detection, face attack models, etc.) and related resources. This is inspired by Awesome-deep-vision, Awesome-adversarial-machine-learning, Awesome-deep-learning-papers, Awesome-NAS and Awesome-Pruing
Please feel free to pull requests or open an issue to add papers.
Contents
Databases
Name | Release year | Attacks | Modalities | #subjects | #videos |
---|---|---|---|---|---|
Ambient-Flash | 2021 | 2D attacks | VIS, additional light flashing | ||
3DMA | 2019 | 3D Mask | VIS,NIR | 67 genuine + 48masks | 920 |
CASIA-SURF 3D HiFi Mask | 2020 | 3D Mask | VIS | ||
CASIA-SURF 3D Mask, paper | 2020 | 3D Mask | VIS | ||
CUHK MMLab CelebA-Spoof | 2020 | 2D Print, Replay (Derived from CelbeA dataset) | VIS | ||
CASIA-SURF Cross-ethnicity Face Anti-spoofing (CeFA), paper | 2019 | 2D Print, Replay; 3D print, silica gel mask | VIS, Depth, Infrared (IR) | 8 | 192 |
IDIAP WMCA | 2019 | 2D Print, Replay and 3D Rigid Mask, Flexible Mask, Paper Mask attacks | VIS, depth, infrared and thermal | 72 | 6716 |
CASIA-SURF, paper | 2018 | 2D Print/Cut | VIS, Depth, Infrared (IR) | 1000 | 21000 |
IDIAP CSMAD, paper | 2018 | Custom Silicone Mask Attack | VIS, near-infrared (NIR), Thermal from long-wave infrared (LWIR), | ||
ROSE-YOUTU | 2018 | 2D Replay, Print, Paper Mask | VIS | 25 | 4225 |
2019 | 2D Replay, Print & 3D Mask | VIS | |||
MSU SiW (Spoofing in the Wild) | 2018 | 2D Replay, Print | VIS | 165 | 4620 |
HKBU-MARs | 2016 | 3D MASK | VIS | 12 | 1008 |
OULU-NPU | 2017 | 2D Replay, Print | VIS | 55 | 5940 |
IDIAP Multispectral-Spoof Dataset | 2015 | 2D Print | VIS, Near-Infrared | 21 | 4704 images |
IDIAP 3D Mask Attack Dataset (3DMAD), paper | 2013 | 3D Mask | VIS, Depth | 17 | 255 |
IDIAP Replay Attack | 2012 | 2D Replay, Print | VIS | ||
CASIA MFSD (FASD) | 2012 | 2D Replay, Print | VIS | 50 | 600 |
Learning-based methods
Year 2022
Title | Venue | Note |
---|---|---|
Learning Meta Pattern for Face Anti-Spoofing | TIFS 2022 | MetaPattern, Code |
Benchmarking Joint Face Spoofing and Forgery Detection with Visual and Physiological Cues | ArXiv 2022 | DeepFake and FAS |
Adaptive Transformers for Robust Few-shot Cross-domain Face Anti-spoofing | ECCV 2022 | - |
Face Anti-Spoofing Using Transformers with Relation-Aware Mechanism | T-BIOM 2022 | |
Domain Generalization via Shuffled Style Assembly for Face Anti-Spoofing | Preprint 2022 | Probabilistic Modeling |
Robust Face Anti-Spoofing with Dual Probabilistic Modeling | CVPR 2022 | Style transfer |
Feature Generation and Hypothesis Verification for Reliable Face Anti-Spoofing | AAAI 2022 | Hypothesis Verification? |
Year 2021
Year 2020
Year 2019-2018
Before 2018
Title | Venue | Note |
---|---|---|
Face Anti-Spoofing Using Patch and Depth-Based CNNs | IJCB 2017 | RGB, 2D Attack, pseudo depth |
Face Spoofing Detection Using Colour Texture Analysis | T-IFS 2016 | RGB, 2D Attack, Color LBP |
Spoofing Face Recognition With 3D Masks | T-IFS 2014 | 3D Mask |
Face Spoofing Detection Through Visual Codebooks of Spectral Temporal Cubes | IEEE TIP 2015 | RGB, 2D Attack, |
Person-Specific Face Anti-Spoofing With Subject Domain Adaptation | T-IFS 2015 | RGB, 2D Attack, Person-specific |
Face Spoof Detection with Image Distortion Analysis | T-IFS 2015 | RGB, 2D Attack, IDA |
Using Visual Rhythms for Detecting Video-Based Facial Spoof Attacks | T-IFS 2015 | VIS, 2D Replay |
Face spoofing detection from single images using texture and local shape analysis | IJCB 2011 | RGB, 2D Presentation Attack, LBP |
System / Mobile Applications
Attack
Title | Venue | Note |
---|---|---|
Light Can Hack Your Face! Black-box Backdoor Attack on Face Recognition Systems | ArXiv 2020 | Face Recognition Attack |
Virtual U: Defeating Face Liveness Detection by Building Virtual Models from Your Public Photos | USENIX SS 2018 | VR tech for spoofing |
Notes
- T-PAMI: IEEE Transactions on Pattern Analysis and Machine Intelligence
- T-IFS: IEEE Transactions on Information Forensics and Security
- T-IP: IEEE Transactions on Image Processing
- T-DSC: IEEE Transactions on Cognitive and Developmental Systems
- T-BIOM IEEE Transactions on Biometrics, Behavior, and Identity Science
- CVPR: Conference on Computer Vision and Pattern Recognition
- ICCV: International Conference on Computer Vision
- ECCV: European Conference on Computer Vision (Springer)
- USENIX SS: Proceedings of USENIX Security Symposium
- NDSS: The Network and Distributed System Security Symposium
- Infocom: IEEE International Conference on Computer Communications
- MobiCom: International Conference On Mobile Computing And Networking
- CCS: ACM Computer and Communications Security Conference
- MobiSys: International Conference on Mobile Systems, Applications, and Services
- IJCB: International Joint Conference on Biometrics
- ArXiv: Papers that only appear in ArXiv are regarded not published, but they are collected here for reference only.
If you feel useful about this repo, we would be happy you are also interested in our works and use our papers your work's reference.
@article{cai2020drlfas,
title={DRL-FAS: A Novel Framework Based on Deep Reinforcement Learning for Face Anti-Spoofing},
author={Cai, Rizhao and Li, Haoliang and Wang, Shiqi and Chen, Changsheng and Kot, Alex C},
journal={IEEE Transactions on Information Forensics and Security},
volume={16},
pages={937--951},
year={2020},
publisher={IEEE},
doi={10.1109/TIFS.2020.3026553}
@ARTICLE{cai2022MP, author={Cai, Rizhao and Li, Zhi and Wan, Renjie and Li, Haoliang and Hu, Yongjian and Kot, Alex C.}, journal={IEEE Transactions on Information Forensics and Security}, title={Learning Meta Pattern for Face Anti-Spoofing}, year={2022}, volume={}, number={}, pages={1-1}, doi={10.1109/TIFS.2022.3158551}} }
@article{li2022one, title={One-Class Knowledge Distillation for Face Presentation Attack Detection}, author={Li, Zhi and Cai, Rizhao and Li, Haoliang and Lam, Kwok-Yan and Hu, Yongjian and Kot, Alex C}, journal={IEEE Transactions on Information Forensics and Security}, year={2022}, publisher={IEEE} }