{"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/whats-the-point-semantic-segmentation-with","title":"What's the Point: Semantic Segmentation with Point Supervision","arxiv_id":"1506.02106","date":"2015-06-06","proceeding":null,"authors":["Amy Bearman","Olga Russakovsky","Vittorio Ferrari","Li Fei-Fei"],"abstract":"The semantic image segmentation task presents a trade-off between test time\naccuracy and training-time annotation cost. Detailed per-pixel annotations\nenable training accurate models but are very time-consuming to obtain,\nimage-level class labels are an order of magnitude cheaper but result in less\naccurate models. We take a natural step from image-level annotation towards\nstronger supervision: we ask annotators to point to an object if one exists. We\nincorporate this point supervision along with a novel objectness potential in\nthe training loss function of a CNN model. Experimental results on the PASCAL\nVOC 2012 benchmark reveal that the combined effect of point-level supervision\nand objectness potential yields an improvement of 12.9% mIOU over image-level\nsupervision. Further, we demonstrate that models trained with point-level\nsupervision are more accurate than models trained with image-level,\nsquiggle-level or full supervision given a fixed annotation budget.","url_abs":"http://arxiv.org/abs/1506.02106v5","url_pdf":"http://arxiv.org/pdf/1506.02106v5.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":"whats-the-point-semantic-segmentation-with","repo_url":"https://github.com/abearman/whats-the-point1","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"1d-cnn","method_name":"1D CNN"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1506.02106","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}