{"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/doobnet-deep-object-occlusion-boundary","title":"DOOBNet: Deep Object Occlusion Boundary Detection from an Image","arxiv_id":"1806.03772","date":"2018-06-11","proceeding":null,"authors":["Guoxia Wang","Xiaohui Liang","Frederick W. B. Li"],"abstract":"Object occlusion boundary detection is a fundamental and crucial research\nproblem in computer vision. This is challenging to solve as encountering the\nextreme boundary/non-boundary class imbalance during training an object\nocclusion boundary detector. In this paper, we propose to address this class\nimbalance by up-weighting the loss contribution of false negative and false\npositive examples with our novel Attention Loss function. We also propose a\nunified end-to-end multi-task deep object occlusion boundary detection network\n(DOOBNet) by sharing convolutional features to simultaneously predict object\nboundary and occlusion orientation. DOOBNet adopts an encoder-decoder structure\nwith skip connection in order to automatically learn multi-scale and\nmulti-level features. We significantly surpass the state-of-the-art on the PIOD\ndataset (ODS F-score of .702) and the BSDS ownership dataset (ODS F-score of\n.555), as well as improving the detecting speed to as 0.037s per image on the\nPIOD dataset.","url_abs":"http://arxiv.org/abs/1806.03772v3","url_pdf":"http://arxiv.org/pdf/1806.03772v3.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":"doobnet-deep-object-occlusion-boundary","repo_url":"https://github.com/GuoxiaWang/DOOBNet","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"boundary-detection","task_name":"Boundary Detection"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"object","task_name":"Object"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.03772","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}