/FASR

Code and Review (survey) for some classical and SOTA face anti-spoofing detection (E.g. LBP, ColorLBP, IDA, patch-based-cnn,CDCN)

Primary LanguagePythonMIT LicenseMIT

FASR

Code and Review (survey) for some classical and SOTA face anti-spoofing detection (E.g. LBP, ColorLBP, IDA, patch-based-cnn,CDCN)

DataSet:

  • Single Modality:
    • Replay-Attack
    • CASIA-FASD
    • MSU-MFSD
    • OULU-NPU
    • SIW
    • SIW-M
    • CelebA
  • Multimodality
    • CASIA-SURF
    • CeFA-SURF
    • HQ-WMCA

Proprocessing

  • video to frame
    from lib.processing_utils import video_to_frames
    video_to_frames(video_path,output_path)
  • frame to face
    from lib.processing_utils import frame_to_face
    frame_to_face(frame_path,output_path)
  • extract face with landmarks