{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/affectnet-a-database-for-facial-expression","title":"AffectNet: A Database for Facial Expression, Valence, and Arousal Computing in the Wild","arxiv_id":"1708.03985","date":"2017-08-14","proceeding":null,"authors":["Ali Mollahosseini","Behzad Hasani","Mohammad H. Mahoor"],"abstract":"Automated affective computing in the wild setting is a challenging problem in\ncomputer vision. Existing annotated databases of facial expressions in the wild\nare small and mostly cover discrete emotions (aka the categorical model). There\nare very limited annotated facial databases for affective computing in the\ncontinuous dimensional model (e.g., valence and arousal). To meet this need, we\ncollected, annotated, and prepared for public distribution a new database of\nfacial emotions in the wild (called AffectNet). AffectNet contains more than\n1,000,000 facial images from the Internet by querying three major search\nengines using 1250 emotion related keywords in six different languages. About\nhalf of the retrieved images were manually annotated for the presence of seven\ndiscrete facial expressions and the intensity of valence and arousal. AffectNet\nis by far the largest database of facial expression, valence, and arousal in\nthe wild enabling research in automated facial expression recognition in two\ndifferent emotion models. Two baseline deep neural networks are used to\nclassify images in the categorical model and predict the intensity of valence\nand arousal. Various evaluation metrics show that our deep neural network\nbaselines can perform better than conventional machine learning methods and\noff-the-shelf facial expression recognition systems.","url_abs":"http://arxiv.org/abs/1708.03985v4","url_pdf":"http://arxiv.org/pdf/1708.03985v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"affectnet-a-database-for-facial-expression","repo_url":"https://github.com/jonathangiguere/Emotion_Image_Classifier","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"facial-expression-recognition-1","task_name":"Facial Expression Recognition"},{"task_slug":"facial-expression-recognition","task_name":"Facial Expression Recognition (FER)"}],"methods":[],"datasets_introduced":[{"slug":"affectnet","name":"AffectNet","full_name":"burak yılmaz"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/facial-expression-recognition-on-affectnet-1","task":"Facial Expression Recognition","dataset":"AffectNet","model":"Up-Sampling","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy (7 emotion)":"-"},"uses_additional_data":false},{"leaderboard":"/sota/facial-expression-recognition-on-affectnet","task":"Facial Expression Recognition (FER)","dataset":"AffectNet","model":"Weighted-Loss","rank_in_archive_order":32,"of":50,"metrics":{"Accuracy (7 emotion)":"-","Accuracy (8 emotion)":"58.0"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1708.03985","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}