{"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/fake-it-till-you-make-it-face-analysis-in-the","title":"Fake It Till You Make It: Face analysis in the wild using synthetic data alone","arxiv_id":"2109.15102","date":"2021-09-30","proceeding":"ICCV 2021 10","authors":["Erroll Wood","Tadas Baltrušaitis","Charlie Hewitt","Sebastian Dziadzio","Matthew Johnson","Virginia Estellers","Thomas J. Cashman","Jamie Shotton"],"abstract":"We demonstrate that it is possible to perform face-related computer vision in the wild using synthetic data alone. The community has long enjoyed the benefits of synthesizing training data with graphics, but the domain gap between real and synthetic data has remained a problem, especially for human faces. Researchers have tried to bridge this gap with data mixing, domain adaptation, and domain-adversarial training, but we show that it is possible to synthesize data with minimal domain gap, so that models trained on synthetic data generalize to real in-the-wild datasets. We describe how to combine a procedurally-generated parametric 3D face model with a comprehensive library of hand-crafted assets to render training images with unprecedented realism and diversity. We train machine learning systems for face-related tasks such as landmark localization and face parsing, showing that synthetic data can both match real data in accuracy as well as open up new approaches where manual labelling would be impossible.","url_abs":"https://arxiv.org/abs/2109.15102v2","url_pdf":"https://arxiv.org/pdf/2109.15102v2.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":[],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"face-alignment","task_name":"Face Alignment"},{"task_slug":"face-model","task_name":"Face Model"},{"task_slug":"face-parsing","task_name":"Face Parsing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-alignment-on-300w","task":"Face Alignment","dataset":"300W","model":"FakeIt","rank_in_archive_order":45,"of":48,"metrics":{"NME_inter-ocular (%, Challenge)":"4.86","NME_inter-ocular (%, Common)":"3.09"},"uses_additional_data":true},{"leaderboard":"/sota/face-parsing-on-helen","task":"Face Parsing","dataset":"Helen","model":"UNet (synthetic)","rank_in_archive_order":2,"of":3,"metrics":{"Mean F1":"92"},"uses_additional_data":true},{"leaderboard":"/sota/face-parsing-on-helen","task":"Face Parsing","dataset":"Helen","model":"UNet (real)","rank_in_archive_order":3,"of":3,"metrics":{"Mean F1":"91.6"},"uses_additional_data":false},{"leaderboard":"/sota/face-parsing-on-lapa","task":"Face Parsing","dataset":"LaPa","model":"UNet (real)","rank_in_archive_order":7,"of":11,"metrics":{"Mean F1":"90.9"},"uses_additional_data":false},{"leaderboard":"/sota/face-parsing-on-lapa","task":"Face Parsing","dataset":"LaPa","model":"UNet (synthetic)","rank_in_archive_order":8,"of":11,"metrics":{"Mean F1":"90.1"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2109.15102","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}