{"url":"/dataset/aff-wild2","name":"Aff-Wild2","full_name":null,"description_markdown":"Aff-Wild2 is a large-scale in-the-wild database and an extension of the Aff-Wild dataset for affect recognition. It approximately doubles the number of included video frames and the number of subjects; thus, improving the variability of the included behaviors and of the involved persons. It is the only existing in-the-wild database with annotations for all 3 main behaviour tasks.\r\n\r\nThe Aff-Wild2 is annotated in a per frame basis for the seven basic expressions (i.e., happiness, surprise, anger, disgust, fear, sadness and the neutral state), twelve action units (AUs 1,2,4,6,7,10,12,15,23,24,25, 26) and valence and arousal. In total Aff-Wild2 consists of 564 videos of around 2.8M frames with 554 subjects.  Aff-Wild2 displays a big diversity in terms of subjects' ages, ethnicities and nationalities; it has also great variations and diversities of environments.\r\n\r\n\r\n\r\nSources:\r\n1)  [Expression, affect, action unit recognition: Aff-wild2, multi-task learning and arcface](https://arxiv.org/pdf/1910.04855);\r\n2)  [The 6th affective behavior analysis in-the-wild (abaw) competition](https://openaccess.thecvf.com/content/CVPR2024W/ABAW/papers/Kollias_The_6th_Affective_Behavior_Analysis_In-the-wild_ABAW_Competition_CVPRW_2024_paper.pdf)","description_withheld":null,"homepage":"https://ibug.doc.ic.ac.uk/resources/aff-wild2/","introduced_date":"2019-09-25","introduced_date_note":null,"introduced_by":{"paper":"/paper/expression-affect-action-unit-recognition-aff","title":"Expression, Affect, Action Unit Recognition: Aff-Wild2, Multi-Task Learning and ArcFace","first_author":"Dimitrios Kollias","url":null},"license":null,"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Emotion Recognition","url":"/task/emotion-recognition","datasets_with_task":"/datasets/task/emotion-recognition"},{"name":"Facial Expression Recognition (FER)","url":"/task/facial-expression-recognition","datasets_with_task":"/datasets/task/facial-expression-recognition"},{"name":"Multi-Task Learning","url":"/task/multi-task-learning","datasets_with_task":"/datasets/task/multi-task-learning"},{"name":"Facial Expression Recognition","url":"/task/facial-expression-recognition-1","datasets_with_task":"/datasets/task/facial-expression-recognition-1"}],"languages":[],"variants":["Aff-Wild2"],"data_loaders":[],"num_papers_in_archive":142,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/facial-expression-recognition-on-aff-wild2","task":"Facial Expression Recognition (FER)","dataset_variant":"Aff-Wild2","rows":2,"metrics":["Accuracy","UAR"],"first_row_in_archive_order":{"model":"GReFEL","paper":"/paper/grefel-geometry-aware-reliable-facial","metrics":{"Accuracy":"72.48"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/facial-expression-recognition-on-aff-wild2-1","task":"Facial Expression Recognition","dataset_variant":"Aff-Wild2","rows":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"ARBEx","paper":"/paper/arbex-attentive-feature-extraction-with","metrics":{"Accuracy":"72.48"},"code_links":[{"title":"takihasan/arbex","url":"https://github.com/takihasan/arbex"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/grefel-geometry-aware-reliable-facial","title":"GReFEL: Geometry-Aware Reliable Facial Expression Learning under Bias and Imbalanced Data Distribution","date":"2024-10-21","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/arbex-attentive-feature-extraction-with","title":"ARBEx: Attentive Feature Extraction with Reliability Balancing for Robust Facial Expression Learning","date":"2023-05-02","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/in-search-of-a-robust-facial-expressions","title":"In Search of a Robust Facial Expressions Recognition Model: A Large-Scale Visual Cross-Corpus Study","date":"2022-10-07","rows_on_this_dataset":1,"code_links":1,"syntology":null}],"syntology_totals":{"read_at":"2026-09-25T09:33:49+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"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."}