{"url":"/dataset/faceforensics","name":"FaceForensics","full_name":null,"description_markdown":"FaceForensics is a video dataset consisting of more than 500,000 frames containing faces from 1004 videos that can be used to study image or video forgeries. All videos are downloaded from Youtube and are cut down to short continuous clips that contain mostly frontal faces. This dataset has two versions:\r\n\r\n* Source-to-Target: where the authors reenact over 1000 videos with new facial expressions extracted from other videos, which e.g. can be used to train a classifier to detect fake images or videos.\r\n\r\n* Selfreenactment: where the authors use Face2Face to reenact the facial expressions of videos with their own facial expressions as input to get pairs of videos, which e.g. can be used to train supervised generative refinement models.","description_withheld":null,"homepage":"http://niessnerlab.org/projects/roessler2018faceforensics.html","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/faceforensics-a-large-scale-video-dataset-for","title":"FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces","first_author":"Andreas Rössler","url":null},"license":{"name":"FaceForenics Terms of Use","url":"http://kaldir.vc.in.tum.de/faceforensics/webpage/FaceForensics_TOS.pdf"},"modalities":[{"name":"Videos","url":"/datasets/modality/videos"}],"tasks":[{"name":"DeepFake Detection","url":"/task/deepfake-detection","datasets_with_task":"/datasets/task/deepfake-detection"}],"languages":[],"variants":["FaceForensics"],"data_loaders":[],"num_papers_in_archive":88,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/deepfake-detection-on-faceforensics","task":"DeepFake Detection","dataset_variant":"FaceForensics","rows":1,"metrics":["DF","FS","FSF","NT","Real","Total Accuracy"],"first_row_in_archive_order":{"model":"XceptionNet","paper":"/paper/faceforensics-learning-to-detect-manipulated","metrics":{"DF":"96.36","FS":"90.29","FSF":"86.86","NT":"80.67","Real":"52.4","Total Accuracy":"70.1"},"code_links":[{"title":"ondyari/FaceForensics","url":"https://github.com/ondyari/FaceForensics"},{"title":"polimi-ispl/icpr2020dfdc","url":"https://github.com/polimi-ispl/icpr2020dfdc"},{"title":"yyk-wew/F3Net","url":"https://github.com/yyk-wew/F3Net"},{"title":"FaceOnLive/DeepFake-Detection-SDK-Linux","url":"https://github.com/FaceOnLive/DeepFake-Detection-SDK-Linux"},{"title":"Baukebrenninkmeijer/HackGAN","url":"https://github.com/Baukebrenninkmeijer/HackGAN"},{"title":"tamlhp/dfd_benchmark","url":"https://github.com/tamlhp/dfd_benchmark"},{"title":"pothabattulasantosh/Detection-of-facial-Manipulated-videos","url":"https://github.com/pothabattulasantosh/Detection-of-facial-Manipulated-videos"},{"title":"pothabattulasantosh/Detection-of-face-Manipulated-videos","url":"https://github.com/pothabattulasantosh/Detection-of-face-Manipulated-videos"},{"title":"SuyashSonawane/fakedetector","url":"https://github.com/SuyashSonawane/fakedetector"},{"title":"DataScienceNigeria/Fake-Detection-dataset-for-deepfake-from-Google-and-Jigsaw","url":"https://github.com/DataScienceNigeria/Fake-Detection-dataset-for-deepfake-from-Google-and-Jigsaw"},{"title":"flynn-chen/faceforensics_benchmark","url":"https://github.com/flynn-chen/faceforensics_benchmark"},{"title":"twyunting/Deepfake_Video_Classifier","url":"https://github.com/twyunting/Deepfake_Video_Classifier"},{"title":"seongilp/DFE604-2020F-FinalProject","url":"https://github.com/seongilp/DFE604-2020F-FinalProject"},{"title":"jhchang/DFDC","url":"https://github.com/jhchang/DFDC"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/faceforensics-learning-to-detect-manipulated","title":"FaceForensics++: Learning to Detect Manipulated Facial Images","date":"2019-01-25","rows_on_this_dataset":1,"code_links":14,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":9,"samples_ran":4,"samples_unverified":5,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":1,"samples_harvested":9,"samples_ran":4,"samples_unverified":5,"pointer_only_for_licence":4,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}