{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/sats-self-attention-transfer-for-continual","title":"SATS: Self-Attention Transfer for Continual Semantic Segmentation","arxiv_id":"2203.07667","date":"2022-03-15","proceeding":null,"authors":["Yiqiao Qiu","Yixing Shen","Zhuohao Sun","Yanchong Zheng","Xiaobin Chang","Weishi Zheng","Ruixuan Wang"],"abstract":"Continually learning to segment more and more types of image regions is a desired capability for many intelligent systems. However, such continual semantic segmentation suffers from the same catastrophic forgetting issue as in continual classification learning. While multiple knowledge distillation strategies originally for continual classification have been well adapted to continual semantic segmentation, they only consider transferring old knowledge based on the outputs from one or more layers of deep fully convolutional networks. Different from existing solutions, this study proposes to transfer a new type of information relevant to knowledge, i.e. the relationships between elements (Eg. pixels or small local regions) within each image which can capture both within-class and between-class knowledge. The relationship information can be effectively obtained from the self-attention maps in a Transformer-style segmentation model. Considering that pixels belonging to the same class in each image often share similar visual properties, a class-specific region pooling is applied to provide more efficient relationship information for knowledge transfer. Extensive evaluations on multiple public benchmarks support that the proposed self-attention transfer method can further effectively alleviate the catastrophic forgetting issue, and its flexible combination with one or more widely adopted strategies significantly outperforms state-of-the-art solutions.","url_abs":"https://arxiv.org/abs/2203.07667v3","url_pdf":"https://arxiv.org/pdf/2203.07667v3.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":"sats-self-attention-transfer-for-continual","repo_url":"https://github.com/QIU023/SATS_Continual_Semantic_Seg","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"continual-semantic-segmentation","task_name":"Continual Semantic Segmentation"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"overlapped-10-1","task_name":"Overlapped 10-1"},{"task_slug":"overlapped-100-10","task_name":"Overlapped 100-10"},{"task_slug":"overlapped-15-1","task_name":"Overlapped 15-1"},{"task_slug":"overlapped-15-5","task_name":"Overlapped 15-5"},{"task_slug":"overlapped-25-25","task_name":"Overlapped 25-25"},{"task_slug":"overlapped-5-3","task_name":"Overlapped 5-3"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/overlapped-10-1-on-pascal-voc-2012","task":"Overlapped 10-1","dataset":"PASCAL VOC 2012","model":"SATS-M","rank_in_archive_order":2,"of":13,"metrics":{"mIoU":"69.27"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-10-1-on-pascal-voc-2012","task":"Overlapped 10-1","dataset":"PASCAL VOC 2012","model":"SATS","rank_in_archive_order":4,"of":13,"metrics":{"mIoU":"61.6"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-100-10-on-ade20k","task":"Overlapped 100-10","dataset":"ADE20K","model":"SATS-M","rank_in_archive_order":2,"of":6,"metrics":{"Mean IoU (test) ":"35.45"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-15-1-on-pascal-voc-2012","task":"Overlapped 15-1","dataset":"PASCAL VOC 2012","model":"SATS-M","rank_in_archive_order":2,"of":13,"metrics":{"mIoU":"76.61"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-15-1-on-pascal-voc-2012","task":"Overlapped 15-1","dataset":"PASCAL VOC 2012","model":"SATS","rank_in_archive_order":3,"of":13,"metrics":{"mIoU":"74.48"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-15-5-on-pascal-voc-2012","task":"Overlapped 15-5","dataset":"PASCAL VOC 2012","model":"SATS-M","rank_in_archive_order":2,"of":13,"metrics":{"Mean IoU (val)":"78.72"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-15-5-on-pascal-voc-2012","task":"Overlapped 15-5","dataset":"PASCAL VOC 2012","model":"SATS","rank_in_archive_order":3,"of":13,"metrics":{"Mean IoU (val)":"75.70"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-25-25-on-ade20k","task":"Overlapped 25-25","dataset":"ADE20K","model":"SATS-M","rank_in_archive_order":1,"of":1,"metrics":{"Mean IoU (test)":"32.56"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-5-3-on-pascal-voc-2012","task":"Overlapped 5-3","dataset":"PASCAL VOC 2012","model":"SATS-M","rank_in_archive_order":2,"of":4,"metrics":{"Mean IoU (test)":"71.36"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-5-3-on-pascal-voc-2012","task":"Overlapped 5-3","dataset":"PASCAL VOC 2012","model":"SATS","rank_in_archive_order":3,"of":4,"metrics":{"Mean IoU (test)":"67.36"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2203.07667","atlas_url":"https://app.syntology.ai/?focus=2203.07667","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2203.07667"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2203.07667\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2203.07667"},"call":"get_concepts_for_paper(arxiv_id=\"2203.07667\")"}],"by_repo":[{"repo":"QIU023/SATS_Continual_Semantic_Seg","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"a0179ebec1624024a51ae71ff04c9b1251ff4dcb89bb0ecc415babc25e8f954e","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}