{"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/siclope-silhouette-based-clothed-people","title":"SiCloPe: Silhouette-Based Clothed People","arxiv_id":"1901.00049","date":"2018-12-31","proceeding":"CVPR 2019 6","authors":["Ryota Natsume","Shunsuke Saito","Zeng Huang","Weikai Chen","Chongyang Ma","Hao Li","Shigeo Morishima"],"abstract":"We introduce a new silhouette-based representation for modeling clothed human\nbodies using deep generative models. Our method can reconstruct a complete and\ntextured 3D model of a person wearing clothes from a single input picture.\nInspired by the visual hull algorithm, our implicit representation uses 2D\nsilhouettes and 3D joints of a body pose to describe the immense shape\ncomplexity and variations of clothed people. Given a segmented 2D silhouette of\na person and its inferred 3D joints from the input picture, we first synthesize\nconsistent silhouettes from novel view points around the subject. The\nsynthesized silhouettes which are the most consistent with the input\nsegmentation are fed into a deep visual hull algorithm for robust 3D shape\nprediction. We then infer the texture of the subject's back view using the\nfrontal image and segmentation mask as input to a conditional generative\nadversarial network. Our experiments demonstrate that our silhouette-based\nmodel is an effective representation and the appearance of the back view can be\npredicted reliably using an image-to-image translation network. While classic\nmethods based on parametric models often fail for single-view images of\nsubjects with challenging clothing, our approach can still produce successful\nresults, which are comparable to those obtained from multi-view input.","url_abs":"http://arxiv.org/abs/1901.00049v2","url_pdf":"http://arxiv.org/pdf/1901.00049v2.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":"siclope-silhouette-based-clothed-people","repo_url":"https://github.com/kuangzijian/drifu-for-animals","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"image-to-image-translation","task_name":"Image-to-Image Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1901.00049","atlas_url":"https://app.syntology.ai/?focus=1901.00049","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}