{"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/segan-segmenting-and-generating-the-invisible","title":"SeGAN: Segmenting and Generating the Invisible","arxiv_id":"1703.10239","date":"2017-03-29","proceeding":"CVPR 2018 6","authors":["Kiana Ehsani","Roozbeh Mottaghi","Ali Farhadi"],"abstract":"Objects often occlude each other in scenes; Inferring their appearance beyond\ntheir visible parts plays an important role in scene understanding, depth\nestimation, object interaction and manipulation. In this paper, we study the\nchallenging problem of completing the appearance of occluded objects. Doing so\nrequires knowing which pixels to paint (segmenting the invisible parts of\nobjects) and what color to paint them (generating the invisible parts). Our\nproposed novel solution, SeGAN, jointly optimizes for both segmentation and\ngeneration of the invisible parts of objects. Our experimental results show\nthat: (a) SeGAN can learn to generate the appearance of the occluded parts of\nobjects; (b) SeGAN outperforms state-of-the-art segmentation baselines for the\ninvisible parts of objects; (c) trained on synthetic photo realistic images,\nSeGAN can reliably segment natural images; (d) by reasoning about occluder\noccludee relations, our method can infer depth layering.","url_abs":"http://arxiv.org/abs/1703.10239v3","url_pdf":"http://arxiv.org/pdf/1703.10239v3.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":"segan-segmenting-and-generating-the-invisible","repo_url":"https://github.com/ehsanik/SeGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"scene-understanding","task_name":"Scene Understanding"},{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.10239","atlas_url":"https://app.syntology.ai/?focus=1703.10239","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}