{"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/decoupled-deep-neural-network-for-semi","title":"Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation","arxiv_id":"1506.04924","date":"2015-06-16","proceeding":"NeurIPS 2015 12","authors":["Seunghoon Hong","Hyeonwoo Noh","Bohyung Han"],"abstract":"We propose a novel deep neural network architecture for semi-supervised\nsemantic segmentation using heterogeneous annotations. Contrary to existing\napproaches posing semantic segmentation as a single task of region-based\nclassification, our algorithm decouples classification and segmentation, and\nlearns a separate network for each task. In this architecture, labels\nassociated with an image are identified by classification network, and binary\nsegmentation is subsequently performed for each identified label in\nsegmentation network. The decoupled architecture enables us to learn\nclassification and segmentation networks separately based on the training data\nwith image-level and pixel-wise class labels, respectively. It facilitates to\nreduce search space for segmentation effectively by exploiting class-specific\nactivation maps obtained from bridging layers. Our algorithm shows outstanding\nperformance compared to other semi-supervised approaches even with much less\ntraining images with strong annotations in PASCAL VOC dataset.","url_abs":"http://arxiv.org/abs/1506.04924v2","url_pdf":"http://arxiv.org/pdf/1506.04924v2.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":"decoupled-deep-neural-network-for-semi","repo_url":"https://github.com/GoNgXiAoPeNg1/caffeBVLCplus","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"decoupled-deep-neural-network-for-semi","repo_url":"https://github.com/HyeonwooNoh/caffe","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"decoupled-deep-neural-network-for-semi","repo_url":"https://github.com/baucheng/caffeFA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-semantic-segmentation","task_name":"Semi-Supervised Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1506.04924","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}