Ștefan Smeu, Dragoș-Alexandru Boldisor, Dan Oneață and Elisabeta OneațăCircumventing shortcuts in audio-visual deepfake detection datasets with unsupervised learningCVPR, 2025
AV-Deepfake1M(AV1M) Dataset: Follow instructions from AV-Deepfake1M
FakeAVCeleb Dataset: Follow instructions from FakeAVCeleb GitHub repo
AVLips Dataset: Follow… See the full description on the dataset page:
https://huggingface.co/datasets/huahua123313/AVH-Alignv2.