{"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/representation-compensation-networks-for","title":"Representation Compensation Networks for Continual Semantic Segmentation","arxiv_id":"2203.05402","date":"2022-03-10","proceeding":"CVPR 2022 1","authors":["Chang-Bin Zhang","Jia-Wen Xiao","Xialei Liu","Ying-Cong Chen","Ming-Ming Cheng"],"abstract":"In this work, we study the continual semantic segmentation problem, where the deep neural networks are required to incorporate new classes continually without catastrophic forgetting. We propose to use a structural re-parameterization mechanism, named representation compensation (RC) module, to decouple the representation learning of both old and new knowledge. The RC module consists of two dynamically evolved branches with one frozen and one trainable. Besides, we design a pooled cube knowledge distillation strategy on both spatial and channel dimensions to further enhance the plasticity and stability of the model. We conduct experiments on two challenging continual semantic segmentation scenarios, continual class segmentation and continual domain segmentation. Without any extra computational overhead and parameters during inference, our method outperforms state-of-the-art performance. The code is available at \\url{https://github.com/zhangchbin/RCIL}.","url_abs":"https://arxiv.org/abs/2203.05402v1","url_pdf":"https://arxiv.org/pdf/2203.05402v1.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":"representation-compensation-networks-for","repo_url":"https://github.com/zhangchbin/rcil","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"class-incremental-learning","task_name":"Class Incremental Learning"},{"task_slug":"continual-learning","task_name":"Continual Learning"},{"task_slug":"continual-semantic-segmentation","task_name":"Continual Semantic Segmentation"},{"task_slug":"disjoint-10-1","task_name":"Disjoint 10-1"},{"task_slug":"disjoint-15-1","task_name":"Disjoint 15-1"},{"task_slug":"disjoint-15-5","task_name":"Disjoint 15-5"},{"task_slug":"domain-1-1","task_name":"Domain 1-1"},{"task_slug":"domain-11-1","task_name":"Domain 11-1"},{"task_slug":"domain-11-5","task_name":"Domain 11-5"},{"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-100-5","task_name":"Overlapped 100-5"},{"task_slug":"overlapped-100-50","task_name":"Overlapped 100-50"},{"task_slug":"overlapped-15-1","task_name":"Overlapped 15-1"},{"task_slug":"overlapped-15-5","task_name":"Overlapped 15-5"},{"task_slug":"overlapped-50-50","task_name":"Overlapped 50-50"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/disjoint-10-1-on-pascal-voc-2012","task":"Disjoint 10-1","dataset":"PASCAL VOC 2012","model":"RCNet-101","rank_in_archive_order":3,"of":8,"metrics":{"mIoU":"18.2"},"uses_additional_data":false},{"leaderboard":"/sota/disjoint-15-1-on-pascal-voc-2012","task":"Disjoint 15-1","dataset":"PASCAL VOC 2012","model":"RCNet-101","rank_in_archive_order":4,"of":9,"metrics":{"mIoU":"54.7"},"uses_additional_data":false},{"leaderboard":"/sota/disjoint-15-5-on-pascal-voc-2012","task":"Disjoint 15-5","dataset":"PASCAL VOC 2012","model":"RCNet-101","rank_in_archive_order":4,"of":9,"metrics":{"Mean IoU":"67.3"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-10-1-on-pascal-voc-2012","task":"Overlapped 10-1","dataset":"PASCAL VOC 2012","model":"RCNet-101","rank_in_archive_order":8,"of":13,"metrics":{"mIoU":"34.3"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-100-10-on-ade20k","task":"Overlapped 100-10","dataset":"ADE20K","model":"RCNet-101","rank_in_archive_order":4,"of":6,"metrics":{"Mean IoU (test) ":"32.1"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-100-5-on-ade20k","task":"Overlapped 100-5","dataset":"ADE20K","model":"RCNet-101","rank_in_archive_order":5,"of":8,"metrics":{"mIoU":"29.6"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-100-50-on-ade20k","task":"Overlapped 100-50","dataset":"ADE20K","model":"RCNet-101","rank_in_archive_order":3,"of":7,"metrics":{"mIoU":"34.5"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-15-1-on-pascal-voc-2012","task":"Overlapped 15-1","dataset":"PASCAL VOC 2012","model":"RCNet-101","rank_in_archive_order":8,"of":13,"metrics":{"mIoU":"59.4"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-15-5-on-pascal-voc-2012","task":"Overlapped 15-5","dataset":"PASCAL VOC 2012","model":"RCNet-101","rank_in_archive_order":5,"of":13,"metrics":{"Mean IoU (val)":"72.4"},"uses_additional_data":false},{"leaderboard":"/sota/overlapped-50-50-on-ade20k","task":"Overlapped 50-50","dataset":"ADE20K","model":"RCNet-101","rank_in_archive_order":3,"of":7,"metrics":{"mIoU":"32.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2203.05402","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2203.05402"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zhangchbin/rcil","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/zhangchbin/RCIL","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":1},"by_repo_kind":{"official":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"453269dfbeac98dd","entry":"ResidualBlock","repo":"zhangchbin/RCIL","repo_kind":"official","path":"modules/residual.py","file_url":"https://github.com/zhangchbin/RCIL/blob/HEAD/modules/residual.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"453269dfbeac98dd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}