{"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/unimatch-v2-pushing-the-limit-of-semi","title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","arxiv_id":"2410.10777","date":"2024-10-14","proceeding":null,"authors":["Lihe Yang","Zhen Zhao","Hengshuang Zhao"],"abstract":"Semi-supervised semantic segmentation (SSS) aims at learning rich visual knowledge from cheap unlabeled images to enhance semantic segmentation capability. Among recent works, UniMatch improves its precedents tremendously by amplifying the practice of weak-to-strong consistency regularization. Subsequent works typically follow similar pipelines and propose various delicate designs. Despite the achieved progress, strangely, even in this flourishing era of numerous powerful vision models, almost all SSS works are still sticking to 1) using outdated ResNet encoders with small-scale ImageNet-1K pre-training, and 2) evaluation on simple Pascal and Cityscapes datasets. In this work, we argue that, it is necessary to switch the baseline of SSS from ResNet-based encoders to more capable ViT-based encoders (e.g., DINOv2) that are pre-trained on massive data. A simple update on the encoder (even using 2x fewer parameters) can bring more significant improvement than careful method designs. Built on this competitive baseline, we present our upgraded and simplified UniMatch V2, inheriting the core spirit of weak-to-strong consistency from V1, but requiring less training cost and providing consistently better results. Additionally, witnessing the gradually saturated performance on Pascal and Cityscapes, we appeal that we should focus on more challenging benchmarks with complex taxonomy, such as ADE20K and COCO datasets. Code, models, and logs of all reported values, are available at https://github.com/LiheYoung/UniMatch-V2.","url_abs":"https://arxiv.org/abs/2410.10777v2","url_pdf":"https://arxiv.org/pdf/2410.10777v2.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":"unimatch-v2-pushing-the-limit-of-semi","repo_url":"https://github.com/LiheYoung/UniMatch-V2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-semantic-segmentation","task_name":"Semi-Supervised Semantic Segmentation"},{"task_slug":"semi-supervised-change-detection","task_name":"Semi-supervised Change Detection"}],"methods":[{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"focus","method_name":"Focus"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-42","task":"Semi-Supervised Semantic Segmentation","dataset":"ADE20K 1/16 labeled","model":"UniMatch V2","rank_in_archive_order":1,"of":5,"metrics":{"Validation mIoU":"46.7"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-41","task":"Semi-Supervised Semantic Segmentation","dataset":"ADE20K 1/32 labeled","model":"UniMatch V2","rank_in_archive_order":1,"of":5,"metrics":{"Validation mIoU":"45.0"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-2","task":"Semi-Supervised Semantic Segmentation","dataset":"COCO 1/128 labeled","model":"UniMatch V2","rank_in_archive_order":1,"of":9,"metrics":{"Validation mIoU":"58.7"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-1","task":"Semi-Supervised Semantic Segmentation","dataset":"COCO 1/256 labeled","model":"UniMatch V2","rank_in_archive_order":1,"of":9,"metrics":{"Validation mIoU":"55.8"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-4","task":"Semi-Supervised Semantic Segmentation","dataset":"COCO 1/32 labeled","model":"UniMatch V2","rank_in_archive_order":1,"of":7,"metrics":{"Validation mIoU":"63.3"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco","task":"Semi-Supervised Semantic Segmentation","dataset":"COCO 1/512 labeled","model":"UniMatch V2","rank_in_archive_order":2,"of":8,"metrics":{"Validation mIoU":"47.9"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-3","task":"Semi-Supervised Semantic Segmentation","dataset":"COCO 1/64 labeled","model":"UniMatch V2","rank_in_archive_order":1,"of":9,"metrics":{"Validation mIoU":"60.4"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-2","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 12.5% labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":33,"metrics":{"Validation mIoU":"84.3%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-1","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 25% labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":30,"metrics":{"Validation mIoU":"84.5%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-8","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 50% labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":23,"metrics":{"Validation mIoU":"85.1%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-22","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 6.25% labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":18,"metrics":{"Validation mIoU":"83.6"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-10","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 1464 labels","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":17,"metrics":{"Validation mIoU":"90.8"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-28","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 183 labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":16,"metrics":{"Validation mIoU":"87.9"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-29","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 366 labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":15,"metrics":{"Validation mIoU":"88.9"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-30","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 732 labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":1,"of":16,"metrics":{"Validation mIoU":"90.0"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-27","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 92 labeled","model":"UniMatch V2 (DINOv2-B)","rank_in_archive_order":2,"of":17,"metrics":{"Validation mIoU":"86.3"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-levir-cd-1","task":"Semi-supervised Change Detection","dataset":"LEVIR-CD - 10% labeled data","model":"UniMatch V2","rank_in_archive_order":1,"of":5,"metrics":{"IoU":"83.8","OA":"99.11"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-levir-cd-2","task":"Semi-supervised Change Detection","dataset":"LEVIR-CD - 20% labeled data","model":"UniMatch V2","rank_in_archive_order":1,"of":4,"metrics":{"IoU":"84.3","OA":"99.14"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-levir-cd-3","task":"Semi-supervised Change Detection","dataset":"LEVIR-CD - 40% labeled data","model":"UniMatch V2","rank_in_archive_order":2,"of":4,"metrics":{"IoU":"84.3","OA":"99.14"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-levir-cd","task":"Semi-supervised Change Detection","dataset":"LEVIR-CD - 5% labeled data","model":"UniMatch V2","rank_in_archive_order":1,"of":5,"metrics":{"IoU":"83.3","OA":"99.08"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-whu-20","task":"Semi-supervised Change Detection","dataset":"WHU - 20% labeled data","model":"UniMatch V2","rank_in_archive_order":1,"of":4,"metrics":{"IoU":"87.9","OA":"99.50"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-change-detection-on-whu-40","task":"Semi-supervised Change Detection","dataset":"WHU - 40% labeled data","model":"UniMatch V2","rank_in_archive_order":1,"of":4,"metrics":{"IoU":"88.6","OA":"99.52"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2410.10777","atlas_url":"https://app.syntology.ai/?focus=2410.10777","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.10777"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/LiheYoung/UniMatch-V2","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"ran":1,"unverified":3},"by_repo_kind":{"official":{"samples":5,"ran":2,"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":"e7fcb6d9ac9deaf4","entry":"named_apply","repo":"LiheYoung/UniMatch-V2","repo_kind":"official","path":"model/backbone/dinov2.py","file_url":"https://github.com/LiheYoung/UniMatch-V2/blob/HEAD/model/backbone/dinov2.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e7fcb6d9ac9deaf4"}},{"code_sha256_prefix":"7c60f8e4d3bbd46f","entry":"setup_distributed","repo":"LiheYoung/UniMatch-V2","repo_kind":"official","path":"util/dist_helper.py","file_url":"https://github.com/LiheYoung/UniMatch-V2/blob/HEAD/util/dist_helper.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7c60f8e4d3bbd46f"}},{"code_sha256_prefix":"21e12abf73350be0","entry":"crop","repo":"LiheYoung/UniMatch-V2","repo_kind":"official","path":"dataset/transform.py","file_url":"https://github.com/LiheYoung/UniMatch-V2/blob/HEAD/dataset/transform.py","link_basis":"plan_row","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":"21e12abf73350be0"}},{"code_sha256_prefix":"9e96d1dddb1f71af","entry":"hflip","repo":"LiheYoung/UniMatch-V2","repo_kind":"official","path":"dataset/transform.py","file_url":"https://github.com/LiheYoung/UniMatch-V2/blob/HEAD/dataset/transform.py","link_basis":"plan_row","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":"9e96d1dddb1f71af"}},{"code_sha256_prefix":"5717ec3a3f33f0a8","entry":"normalize","repo":"LiheYoung/UniMatch-V2","repo_kind":"official","path":"dataset/transform.py","file_url":"https://github.com/LiheYoung/UniMatch-V2/blob/HEAD/dataset/transform.py","link_basis":"plan_row","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":"5717ec3a3f33f0a8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}