{"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/deepflux-for-skeletons-in-the-wild","title":"DeepFlux for Skeletons in the Wild","arxiv_id":"1811.12608","date":"2018-11-30","proceeding":"CVPR 2019 6","authors":["Yukang Wang","Yongchao Xu","Stavros Tsogkas","Xiang Bai","Sven Dickinson","Kaleem Siddiqi"],"abstract":"Computing object skeletons in natural images is challenging, owing to large\nvariations in object appearance and scale, and the complexity of handling\nbackground clutter. Many recent methods frame object skeleton detection as a\nbinary pixel classification problem, which is similar in spirit to\nlearning-based edge detection, as well as to semantic segmentation methods. In\nthe present article, we depart from this strategy by training a CNN to predict\na two-dimensional vector field, which maps each scene point to a candidate\nskeleton pixel, in the spirit of flux-based skeletonization algorithms. This\n\"image context flux\" representation has two major advantages over previous\napproaches. First, it explicitly encodes the relative position of skeletal\npixels to semantically meaningful entities, such as the image points in their\nspatial context, and hence also the implied object boundaries. Second, since\nthe skeleton detection context is a region-based vector field, it is better\nable to cope with object parts of large width. We evaluate the proposed method\non three benchmark datasets for skeleton detection and two for symmetry\ndetection, achieving consistently superior performance over state-of-the-art\nmethods.","url_abs":"http://arxiv.org/abs/1811.12608v1","url_pdf":"http://arxiv.org/pdf/1811.12608v1.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":"deepflux-for-skeletons-in-the-wild","repo_url":"https://github.com/YukangWang/DeepFlux","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"deepflux-for-skeletons-in-the-wild","repo_url":"https://dagshub.com/Bharat-mtr/DeepFlux","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"edge-detection","task_name":"Edge Detection"},{"task_slug":"object","task_name":"Object"},{"task_slug":"object-skeleton-detection","task_name":"Object Skeleton Detection"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"symmetry-detection","task_name":"Symmetry Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/object-skeleton-detection-on-sk-large","task":"Object Skeleton Detection","dataset":"SK-LARGE","model":"DeepFlux","rank_in_archive_order":1,"of":2,"metrics":{"F-Measure":"0.732"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1811.12608","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}