Skip to content

Latest commit

 

History

History
55 lines (47 loc) · 2.59 KB

README.md

File metadata and controls

55 lines (47 loc) · 2.59 KB

MetaPattern_FAS

The code repo for the paper "Learning Meta Pattern for Face Anti-Spoofing", IEEE TIFS 2022 / (ArXiv version).

im

Steps for reproduction

Data extraction

  1. Please download the datasets you need. Due to the license, I cannot redistribute the datasets. You have to download the datasets by submitting your requests. The links for downloading the datasets can be found in https://github.com/RizhaoCai/Awesome-FAS.
  2. Configure directories
    2.1 Go to data/extract_frames_mtcnn.py
    2.2 Configure the MTCCN path (check line 14 of data/extract_frames_mtcnn.py)
    2.3 Check lines 22 and 23. Configure ROOT_DIR and SAVE_DIR
  3. Run scripts to extract data frames 
    cd data
# Change the DATASET_NAME you need; Check the keys of the dict 'DATASET_DIR' in data/extract_frames_mtcnn.py
DATASET_NAME='CASIA-FASD'; python extract_frames_mtcnn.py ${DATASET_NAME}
cd ..
  4. Configure default configs in config/defaults.py
    This file provide many basic configs for data, network, and training. Some model-specific/custom configs are in each sub-folders of [models] (e.g. models/HFN_MP/custom_config.py). Custom configs will be merged with the deafult config.

Please configure _C.DATA.ROOT_DIR in config/defaults.py, which is the root directory where you save the preprocessed data.

Run training

# Train models for the C+I+O to M protocol
python train.py --trainer HFN_MP --config models/HFN_MP/CIO-M.yaml OUTPUT_DIR output/HFN_MP/CIO-M/NF2_align_InnerLoops4 TRAIN.NUM_FRAMES 2 TEST.NUM_FRAMES 2 TRAIN.INNER_LOOPS 4 TRAIN.META_LEARNING_RATE 0.001 TRAIN.IMAGENET_PRETRAIN True

Citation

If you use this repo, please kindly cite the below papers

@article{cai2022learning,
  title={Learning Meta Pattern for Face Anti-Spoofing},
  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},
  volume={17},
  pages={1201--1213},
  year={2022},
  publisher={IEEE}
}


@article{cai2020drl,
  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}
}