{"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/region-based-semantic-segmentation-with-end","title":"Region-based semantic segmentation with end-to-end training","arxiv_id":"1607.07671","date":"2016-07-26","proceeding":null,"authors":["Holger Caesar","Jasper Uijlings","Vittorio Ferrari"],"abstract":"We propose a novel method for semantic segmentation, the task of labeling\neach pixel in an image with a semantic class. Our method combines the\nadvantages of the two main competing paradigms. Methods based on region\nclassification offer proper spatial support for appearance measurements, but\ntypically operate in two separate stages, none of which targets pixel labeling\nperformance at the end of the pipeline. More recent fully convolutional methods\nare capable of end-to-end training for the final pixel labeling, but resort to\nfixed patches as spatial support. We show how to modify modern region-based\napproaches to enable end-to-end training for semantic segmentation. This is\nachieved via a differentiable region-to-pixel layer and a differentiable\nfree-form Region-of-Interest pooling layer. Our method improves the\nstate-of-the-art in terms of class-average accuracy with 64.0% on SIFT Flow and\n49.9% on PASCAL Context, and is particularly accurate at object boundaries.","url_abs":"http://arxiv.org/abs/1607.07671v1","url_pdf":"http://arxiv.org/pdf/1607.07671v1.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":"region-based-semantic-segmentation-with-end","repo_url":"https://github.com/nightrome/matconvnet-calvin","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-pascal-context","task":"Semantic Segmentation","dataset":"PASCAL Context","model":"RBE2E","rank_in_archive_order":65,"of":66,"metrics":{"Mean Accuracy":"49.9","Pixel Accuracy":"62.4","mIoU":"32.5"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-sift-flow","task":"Semantic Segmentation","dataset":"SIFT-flow","model":"RBE2E","rank_in_archive_order":1,"of":3,"metrics":{"Mean Accuracy":"64","Pixel Accuracy":"84.3"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}