{"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/extreme-3d-face-reconstruction-seeing-through","title":"Extreme 3D Face Reconstruction: Seeing Through Occlusions","arxiv_id":"1712.05083","date":"2017-12-14","proceeding":"CVPR 2018 6","authors":["Anh Tuan Tran","Tal Hassner","Iacopo Masi","Eran Paz","Yuval Nirkin","Gerard Medioni"],"abstract":"Existing single view, 3D face reconstruction methods can produce beautifully\ndetailed 3D results, but typically only for near frontal, unobstructed\nviewpoints. We describe a system designed to provide detailed 3D\nreconstructions of faces viewed under extreme conditions, out of plane\nrotations, and occlusions. Motivated by the concept of bump mapping, we propose\na layered approach which decouples estimation of a global shape from its\nmid-level details (e.g., wrinkles). We estimate a coarse 3D face shape which\nacts as a foundation and then separately layer this foundation with details\nrepresented by a bump map. We show how a deep convolutional encoder-decoder can\nbe used to estimate such bump maps. We further show how this approach naturally\nextends to generate plausible details for occluded facial regions. We test our\napproach and its components extensively, quantitatively demonstrating the\ninvariance of our estimated facial details. We further provide numerous\nqualitative examples showing that our method produces detailed 3D face shapes\nin viewing conditions where existing state of the art often break down.","url_abs":"http://arxiv.org/abs/1712.05083v2","url_pdf":"http://arxiv.org/pdf/1712.05083v2.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":"extreme-3d-face-reconstruction-seeing-through","repo_url":"https://github.com/anhttran/extreme_3d_faces","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-face-reconstruction","task_name":"3D Face Reconstruction"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"face-reconstruction","task_name":"Face Reconstruction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}