{"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/semantic-segmentation-with-reverse-attention","title":"Semantic Segmentation with Reverse Attention","arxiv_id":"1707.06426","date":"2017-07-20","proceeding":null,"authors":["Qin Huang","Chunyang Xia","Chi-Hao Wu","Siyang Li","Ye Wang","Yuhang Song","C. -C. Jay Kuo"],"abstract":"Recent development in fully convolutional neural network enables efficient\nend-to-end learning of semantic segmentation. Traditionally, the convolutional\nclassifiers are taught to learn the representative semantic features of labeled\nsemantic objects. In this work, we propose a reverse attention network (RAN)\narchitecture that trains the network to capture the opposite concept (i.e.,\nwhat are not associated with a target class) as well. The RAN is a three-branch\nnetwork that performs the direct, reverse and reverse-attention learning\nprocesses simultaneously. Extensive experiments are conducted to show the\neffectiveness of the RAN in semantic segmentation. Being built upon the\nDeepLabv2-LargeFOV, the RAN achieves the state-of-the-art mIoU score (48.1%)\nfor the challenging PASCAL-Context dataset. Significant performance\nimprovements are also observed for the PASCAL-VOC, Person-Part, NYUDv2 and\nADE20K datasets.","url_abs":"http://arxiv.org/abs/1707.06426v1","url_pdf":"http://arxiv.org/pdf/1707.06426v1.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":[],"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-nyu-depth-v2","task":"Semantic Segmentation","dataset":"NYU Depth v2","model":"RAN","rank_in_archive_order":108,"of":121,"metrics":{"Mean IoU":"41.2%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.06426","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}