{"url":"/task/semi-supervised-semantic-segmentation","name":"Semi-Supervised Semantic Segmentation","slug":"semi-supervised-semantic-segmentation","description_markdown":"Models that are trained with a small number of labeled examples and a large number of unlabeled examples and whose aim is to learn to segment an image (i.e. assign a class to every pixel).","categories":[{"name":"Computer Code","url":"/area/computer-code"},{"name":"Computer Vision","url":"/area/computer-vision"},{"name":"Medical","url":"/area/medical"},{"name":"Robots","url":"/area/robots"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":190,"papers_with_code":109,"benchmarks":45,"benchmark_tables_in_archive":45,"benchmark_tables_shown":45,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":13,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-4","slug":"semi-supervised-semantic-segmentation-on-4","dataset":"Pascal VOC 2012 12.5% labeled","dataset_url":null,"rows_in_archive":38,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"SemiOVS (w/ UniMatch, ResNet-101)","paper_title":"Leveraging Out-of-Distribution Unlabeled Images: Semi-Supervised Semantic Segmentation with an Open-Vocabulary Model","paper_url":"/paper/leveraging-out-of-distribution-unlabeled","paper_date":"2025-07-04","arxiv_id":"2507.03302","code_links":[{"title":"wooseok-shin/SemiOVS","url":"https://github.com/wooseok-shin/SemiOVS"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-2","slug":"semi-supervised-semantic-segmentation-on-2","dataset":"Cityscapes 12.5% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":33,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-1","slug":"semi-supervised-semantic-segmentation-on-1","dataset":"Cityscapes 25% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":30,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-9","slug":"semi-supervised-semantic-segmentation-on-9","dataset":"PASCAL VOC 2012 25% labeled","dataset_url":null,"rows_in_archive":27,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"SemiOVS (w/ PrevMatch, ResNet-101)","paper_title":"Leveraging Out-of-Distribution Unlabeled Images: Semi-Supervised Semantic Segmentation with an Open-Vocabulary Model","paper_url":"/paper/leveraging-out-of-distribution-unlabeled","paper_date":"2025-07-04","arxiv_id":"2507.03302","code_links":[{"title":"wooseok-shin/SemiOVS","url":"https://github.com/wooseok-shin/SemiOVS"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-8","slug":"semi-supervised-semantic-segmentation-on-8","dataset":"Cityscapes 50% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":23,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-21","slug":"semi-supervised-semantic-segmentation-on-21","dataset":"Pascal VOC 2012 6.25% labeled","dataset_url":null,"rows_in_archive":19,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"SemiOVS (w/ PrevMatch, ResNet-101)","paper_title":"Leveraging Out-of-Distribution Unlabeled Images: Semi-Supervised Semantic Segmentation with an Open-Vocabulary Model","paper_url":"/paper/leveraging-out-of-distribution-unlabeled","paper_date":"2025-07-04","arxiv_id":"2507.03302","code_links":[{"title":"wooseok-shin/SemiOVS","url":"https://github.com/wooseok-shin/SemiOVS"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-22","slug":"semi-supervised-semantic-segmentation-on-22","dataset":"Cityscapes 6.25% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":18,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-10","slug":"semi-supervised-semantic-segmentation-on-10","dataset":"PASCAL VOC 2012 1464 labels","dataset_url":null,"rows_in_archive":17,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-27","slug":"semi-supervised-semantic-segmentation-on-27","dataset":"PASCAL VOC 2012 92 labeled","dataset_url":null,"rows_in_archive":17,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"SemiOVS (w/ SemiVL, ViT-B/16)","paper_title":"Leveraging Out-of-Distribution Unlabeled Images: Semi-Supervised Semantic Segmentation with an Open-Vocabulary Model","paper_url":"/paper/leveraging-out-of-distribution-unlabeled","paper_date":"2025-07-04","arxiv_id":"2507.03302","code_links":[{"title":"wooseok-shin/SemiOVS","url":"https://github.com/wooseok-shin/SemiOVS"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-28","slug":"semi-supervised-semantic-segmentation-on-28","dataset":"PASCAL VOC 2012 183 labeled","dataset_url":null,"rows_in_archive":16,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-30","slug":"semi-supervised-semantic-segmentation-on-30","dataset":"PASCAL VOC 2012 732 labeled","dataset_url":null,"rows_in_archive":16,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-29","slug":"semi-supervised-semantic-segmentation-on-29","dataset":"PASCAL VOC 2012 366 labeled","dataset_url":null,"rows_in_archive":15,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2 (DINOv2-B)","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-15","slug":"semi-supervised-semantic-segmentation-on-15","dataset":"PASCAL VOC 2012 50%","dataset_url":null,"rows_in_archive":14,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"S4MC","paper_title":"Semi-Supervised Semantic Segmentation via Marginal Contextual Information","paper_url":"/paper/semi-supervised-semantic-segmentation-via-3","paper_date":"2023-08-26","arxiv_id":"2308.13900","code_links":[{"title":"s4mcontext/s4mc","url":"https://github.com/s4mcontext/s4mc"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-5","slug":"semi-supervised-semantic-segmentation-on-5","dataset":"Pascal VOC 2012 5% labeled","dataset_url":null,"rows_in_archive":14,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pretrained)","paper_title":"Bootstrapping Semantic Segmentation with Regional Contrast","paper_url":"/paper/bootstrapping-semantic-segmentation-with","paper_date":"2021-04-09","arxiv_id":"2104.04465","code_links":[{"title":"lorenmt/reco","url":"https://github.com/lorenmt/reco"},{"title":"dbash/zerowaste","url":"https://github.com/dbash/zerowaste"}],"syntology":{"n":6,"n_ran":3,"n_unverified":3,"n_pointer_only":2}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-3","slug":"semi-supervised-semantic-segmentation-on-3","dataset":"Cityscapes 100 samples labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":13,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"SemiVL (ViT-B/16)","paper_title":"SemiVL: Semi-Supervised Semantic Segmentation with Vision-Language Guidance","paper_url":"/paper/semivl-semi-supervised-semantic-segmentation","paper_date":"2023-11-27","arxiv_id":"2311.16241","code_links":[{"title":"google-research/semivl","url":"https://github.com/google-research/semivl"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-24","slug":"semi-supervised-semantic-segmentation-on-24","dataset":"SemanticKITTI","dataset_url":"/dataset/semantickitti","rows_in_archive":12,"metrics":["mIoU (0.5% Labels)","mIoU (1% Labels)","mIoU (2% Labels)","mIoU (5% Labels)","mIoU (10% Labels)","mIoU (20% Labels)","mIoU (50% Labels)","mIOU (1% Test set)"],"first_row_in_archive_order":{"model":"PLE (Voxel)","paper_title":"Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation","paper_url":"/paper/learning-from-spatio-temporal-correlation-for","paper_date":"2024-10-09","arxiv_id":"2410.06893","code_links":[{"title":"halbielee/ple","url":"https://github.com/halbielee/ple"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-6","slug":"semi-supervised-semantic-segmentation-on-6","dataset":"Pascal VOC 2012 2% labeled","dataset_url":null,"rows_in_archive":12,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pretrained)","paper_title":"Bootstrapping Semantic Segmentation with Regional Contrast","paper_url":"/paper/bootstrapping-semantic-segmentation-with","paper_date":"2021-04-09","arxiv_id":"2104.04465","code_links":[{"title":"lorenmt/reco","url":"https://github.com/lorenmt/reco"},{"title":"dbash/zerowaste","url":"https://github.com/dbash/zerowaste"}],"syntology":{"n":6,"n_ran":3,"n_unverified":3,"n_pointer_only":2}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-25","slug":"semi-supervised-semantic-segmentation-on-25","dataset":"nuScenes","dataset_url":"/dataset/nuscenes","rows_in_archive":11,"metrics":["mIoU (0.5% Labels)","mIoU (1% Labels)","mIoU (2% Labels)","mIoU (5% Labels)","mIoU (10% Labels)","mIoU (20% Labels)","mIoU (50% Labels)"],"first_row_in_archive_order":{"model":"PLE (Voxel)","paper_title":"Learning from Spatio-temporal Correlation for Semi-Supervised LiDAR Semantic Segmentation","paper_url":"/paper/learning-from-spatio-temporal-correlation-for","paper_date":"2024-10-09","arxiv_id":"2410.06893","code_links":[{"title":"halbielee/ple","url":"https://github.com/halbielee/ple"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-23","slug":"semi-supervised-semantic-segmentation-on-23","dataset":"ScribbleKITTI","dataset_url":"/dataset/scribblekitti","rows_in_archive":9,"metrics":["mIoU (1% Labels)","mIoU (10% Labels)","mIoU (20% Labels)","mIoU (50% Labels)"],"first_row_in_archive_order":{"model":"LaserMix (Voxel)","paper_title":"LaserMix for Semi-Supervised LiDAR Semantic Segmentation","paper_url":"/paper/lasermix-for-semi-supervised-lidar-semantic","paper_date":"2022-06-30","arxiv_id":"2207.00026","code_links":[{"title":"ldkong1205/LaserMix","url":"https://github.com/ldkong1205/LaserMix"},{"title":"yuan-zm/dgt-st","url":"https://github.com/yuan-zm/dgt-st"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-1","slug":"semi-supervised-semantic-segmentation-on-coco-1","dataset":"COCO 1/256 labeled","dataset_url":null,"rows_in_archive":9,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-2","slug":"semi-supervised-semantic-segmentation-on-coco-2","dataset":"COCO 1/128 labeled","dataset_url":null,"rows_in_archive":9,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-3","slug":"semi-supervised-semantic-segmentation-on-coco-3","dataset":"COCO 1/64 labeled","dataset_url":null,"rows_in_archive":9,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco","slug":"semi-supervised-semantic-segmentation-on-coco","dataset":"COCO 1/512 labeled","dataset_url":null,"rows_in_archive":8,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"SemiVL","paper_title":"SemiVL: Semi-Supervised Semantic Segmentation with Vision-Language Guidance","paper_url":"/paper/semivl-semi-supervised-semantic-segmentation","paper_date":"2023-11-27","arxiv_id":"2311.16241","code_links":[{"title":"google-research/semivl","url":"https://github.com/google-research/semivl"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-coco-4","slug":"semi-supervised-semantic-segmentation-on-coco-4","dataset":"COCO 1/32 labeled","dataset_url":null,"rows_in_archive":7,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-7","slug":"semi-supervised-semantic-segmentation-on-7","dataset":"Pascal VOC 2012 1% labeled","dataset_url":null,"rows_in_archive":6,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"ReCo (DeepLab v3+ with ResNet-101 backbone, ImageNet pre-trained)","paper_title":"Bootstrapping Semantic Segmentation with Regional Contrast","paper_url":"/paper/bootstrapping-semantic-segmentation-with","paper_date":"2021-04-09","arxiv_id":"2104.04465","code_links":[{"title":"lorenmt/reco","url":"https://github.com/lorenmt/reco"},{"title":"dbash/zerowaste","url":"https://github.com/dbash/zerowaste"}],"syntology":{"n":6,"n_ran":3,"n_unverified":3,"n_pointer_only":2}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-41","slug":"semi-supervised-semantic-segmentation-on-41","dataset":"ADE20K 1/32 labeled","dataset_url":"/dataset/ade20k","rows_in_archive":5,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-42","slug":"semi-supervised-semantic-segmentation-on-42","dataset":"ADE20K 1/16 labeled","dataset_url":"/dataset/ade20k","rows_in_archive":5,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"UniMatch V2","paper_title":"UniMatch V2: Pushing the Limit of Semi-Supervised Semantic Segmentation","paper_url":"/paper/unimatch-v2-pushing-the-limit-of-semi","paper_date":"2024-10-14","arxiv_id":"2410.10777","code_links":[{"title":"LiheYoung/UniMatch-V2","url":"https://github.com/LiheYoung/UniMatch-V2"}],"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-18","slug":"semi-supervised-semantic-segmentation-on-18","dataset":"Cityscapes 2% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":3,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"GIST and RIST (DeepLabv2 with ResNet101, MSCOCO pre-trained)","paper_title":"The GIST and RIST of Iterative Self-Training for Semi-Supervised Segmentation","paper_url":"/paper/the-gist-and-rist-of-iterative-self-training","paper_date":"2021-03-31","arxiv_id":"2103.17105","code_links":[],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-19","slug":"semi-supervised-semantic-segmentation-on-19","dataset":"Cityscapes 5% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":3,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"GIST and RIST (DeepLabv2 with ResNet101, MSCOCO pre-trained)","paper_title":"The GIST and RIST of Iterative Self-Training for Semi-Supervised Segmentation","paper_url":"/paper/the-gist-and-rist-of-iterative-self-training","paper_date":"2021-03-31","arxiv_id":"2103.17105","code_links":[],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-35","slug":"semi-supervised-semantic-segmentation-on-35","dataset":"Cityscapes 93 labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":3,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)","paper_title":"Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning","paper_url":"/paper/semi-supervised-semantic-segmentation-via-2","paper_date":"2021-10-11","arxiv_id":"2110.05474","code_links":[{"title":"hzhupku/semiseg-ael","url":"https://github.com/hzhupku/semiseg-ael"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-44","slug":"semi-supervised-semantic-segmentation-on-44","dataset":"Pascal VOC 2012 50% labeled","dataset_url":null,"rows_in_archive":3,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"AllSpark","paper_title":"AllSpark: Reborn Labeled Features from Unlabeled in Transformer for Semi-Supervised Semantic Segmentation","paper_url":"/paper/allspark-reborn-labeled-features-from","paper_date":"2024-03-04","arxiv_id":"2403.01818","code_links":[{"title":"xmed-lab/AllSpark","url":"https://github.com/xmed-lab/AllSpark"}],"syntology":{"n":2,"n_ran":1,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-11","slug":"semi-supervised-semantic-segmentation-on-11","dataset":"PASCAL Context 12.5% labeled","dataset_url":null,"rows_in_archive":2,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","paper_title":"GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference","paper_url":"/paper/guidedmix-net-learning-to-improve-pseudo","paper_date":"2021-06-29","arxiv_id":"2106.15064","code_links":[{"title":"yh-pengtu/GuidedMix-Net","url":"https://github.com/yh-pengtu/GuidedMix-Net"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-12","slug":"semi-supervised-semantic-segmentation-on-12","dataset":"PASCAL Context 25% labeled","dataset_url":null,"rows_in_archive":2,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","paper_title":"GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference","paper_url":"/paper/guidedmix-net-learning-to-improve-pseudo","paper_date":"2021-06-29","arxiv_id":"2106.15064","code_links":[{"title":"yh-pengtu/GuidedMix-Net","url":"https://github.com/yh-pengtu/GuidedMix-Net"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-32","slug":"semi-supervised-semantic-segmentation-on-32","dataset":"Stanford 2D-3D","dataset_url":null,"rows_in_archive":2,"metrics":["mIoU (0.1% labels)","MM-Robust mIoU (0.1% labels)"],"first_row_in_archive_order":{"model":"M3L (Linear Fusion - Segformer B2)","paper_title":"Missing Modality Robustness in Semi-Supervised Multi-Modal Semantic Segmentation","paper_url":"/paper/missing-modality-robustness-in-semi","paper_date":"2023-04-21","arxiv_id":"2304.10756","code_links":[{"title":"harshm121/m3l","url":"https://github.com/harshm121/m3l"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-40","slug":"semi-supervised-semantic-segmentation-on-40","dataset":"Cityscapes with extra (no coarse labels)","dataset_url":"/dataset/cityscapes","rows_in_archive":2,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"Dense FixMatch (DeepLabv3+ ResNet-101, over-sampling, single pass eval)","paper_title":"Dense FixMatch: a simple semi-supervised learning method for pixel-wise prediction tasks","paper_url":"/paper/dense-fixmatch-a-simple-semi-supervised","paper_date":"2022-10-18","arxiv_id":"2210.09919","code_links":[{"title":"miquelmarti/DenseFixMatch","url":"https://github.com/miquelmarti/DenseFixMatch"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-45","slug":"semi-supervised-semantic-segmentation-on-45","dataset":"WoodScape","dataset_url":"/dataset/woodscape","rows_in_archive":2,"metrics":["Mean IoU"],"first_row_in_archive_order":{"model":"FishSegSSL","paper_title":"FishSegSSL: A Semi-Supervised Semantic Segmentation Framework for Fish-Eye Images","paper_url":"/paper/fishsegssl-a-semi-supervised-semantic","paper_date":"2024-03-15","arxiv_id":null,"code_links":[{"title":"snehaputul/FishSegSSL","url":"https://github.com/snehaputul/FishSegSSL"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-16","slug":"semi-supervised-semantic-segmentation-on-16","dataset":"PASCAL VOC 2012 500 labels","dataset_url":null,"rows_in_archive":1,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)","paper_title":"GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference","paper_url":"/paper/guidedmix-net-learning-to-improve-pseudo","paper_date":"2021-06-29","arxiv_id":"2106.15064","code_links":[{"title":"yh-pengtu/GuidedMix-Net","url":"https://github.com/yh-pengtu/GuidedMix-Net"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-17","slug":"semi-supervised-semantic-segmentation-on-17","dataset":"PASCAL VOC 2012 1000 labels","dataset_url":null,"rows_in_archive":1,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)","paper_title":"GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference","paper_url":"/paper/guidedmix-net-learning-to-improve-pseudo","paper_date":"2021-06-29","arxiv_id":"2106.15064","code_links":[{"title":"yh-pengtu/GuidedMix-Net","url":"https://github.com/yh-pengtu/GuidedMix-Net"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-2017","slug":"semi-supervised-semantic-segmentation-on-2017","dataset":"2017 Robotic Instrument Segmentation Challenge","dataset_url":"/dataset/2017-robotic-instrument-segmentation","rows_in_archive":1,"metrics":["DSC"],"first_row_in_archive_order":{"model":"MMS (20% Labeled)","paper_title":"Min-Max Similarity: A Contrastive Semi-Supervised Deep Learning Network for Surgical Tools Segmentation","paper_url":"/paper/min-max-similarity-a-contrastive-learning","paper_date":"2022-03-29","arxiv_id":"2203.15177","code_links":[{"title":"angeloucn/min_max_similarity","url":"https://github.com/angeloucn/min_max_similarity"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-2d","slug":"semi-supervised-semantic-segmentation-on-2d","dataset":"2D-3D-S","dataset_url":"/dataset/2d-3d-s","rows_in_archive":1,"metrics":["mIoU (0.1% labels)","mIoU (0.2% labels)","mIoU (1% labels)"],"first_row_in_archive_order":{"model":"M3L (Linear Fusion B2)","paper_title":"Missing Modality Robustness in Semi-Supervised Multi-Modal Semantic Segmentation","paper_url":"/paper/missing-modality-robustness-in-semi","paper_date":"2023-04-21","arxiv_id":"2304.10756","code_links":[{"title":"harshm121/m3l","url":"https://github.com/harshm121/m3l"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-33","slug":"semi-supervised-semantic-segmentation-on-33","dataset":"Kvasir-Instrument","dataset_url":"/dataset/kvasir-instrument","rows_in_archive":1,"metrics":["Dice (Average)"],"first_row_in_archive_order":{"model":"MMS(20% labeled)","paper_title":"Min-Max Similarity: A Contrastive Semi-Supervised Deep Learning Network for Surgical Tools Segmentation","paper_url":"/paper/min-max-similarity-a-contrastive-learning","paper_date":"2022-03-29","arxiv_id":"2203.15177","code_links":[{"title":"angeloucn/min_max_similarity","url":"https://github.com/angeloucn/min_max_similarity"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-34","slug":"semi-supervised-semantic-segmentation-on-34","dataset":"KiTS19","dataset_url":"/dataset/kits19","rows_in_archive":1,"metrics":["Avg DSC"],"first_row_in_archive_order":{"model":"PatchCL","paper_title":"Pseudo-Label Guided Contrastive Learning for Semi-Supervised Medical Image Segmentation","paper_url":"/paper/pseudo-label-guided-contrastive-learning-for","paper_date":"2023-01-01","arxiv_id":null,"code_links":[{"title":"HiLab-git/SSL4MIS","url":"https://github.com/HiLab-git/SSL4MIS"},{"title":"hritam-98/patchcl-medseg","url":"https://github.com/hritam-98/patchcl-medseg"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-36","slug":"semi-supervised-semantic-segmentation-on-36","dataset":"PASCAL VOC 2012 331 labeled","dataset_url":null,"rows_in_archive":1,"metrics":["Validation mIoU"],"first_row_in_archive_order":{"model":"AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K)","paper_title":"Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning","paper_url":"/paper/semi-supervised-semantic-segmentation-via-2","paper_date":"2021-10-11","arxiv_id":"2110.05474","code_links":[{"title":"hzhupku/semiseg-ael","url":"https://github.com/hzhupku/semiseg-ael"}],"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":0}}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-43","slug":"semi-supervised-semantic-segmentation-on-43","dataset":"Cityscapes 10% labeled","dataset_url":"/dataset/cityscapes","rows_in_archive":1,"metrics":["Mean IoU (class)"],"first_row_in_archive_order":{"model":"IM++ (416x208, 2.7m parameters, no pretraining)","paper_title":"Inconsistency Masks: Removing the Uncertainty from Input-Pseudo-Label Pairs","paper_url":"/paper/inconsistency-masks-removing-the-uncertainty","paper_date":"2024-01-25","arxiv_id":"2401.14387","code_links":[{"title":"michaelvorndran/inconsistencymasks","url":"https://github.com/michaelvorndran/inconsistencymasks"}],"syntology":null}},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-suim","slug":"semi-supervised-semantic-segmentation-on-suim","dataset":"SUIM","dataset_url":"/dataset/suim","rows_in_archive":1,"metrics":["Mean IoU (class)"],"first_row_in_archive_order":{"model":"AIM+ (256x256, 2.7m parameters, 10% labeled data, no pretraining)","paper_title":"Inconsistency Masks: Removing the Uncertainty from Input-Pseudo-Label Pairs","paper_url":"/paper/inconsistency-masks-removing-the-uncertainty","paper_date":"2024-01-25","arxiv_id":"2401.14387","code_links":[{"title":"michaelvorndran/inconsistencymasks","url":"https://github.com/michaelvorndran/inconsistencymasks"}],"syntology":null}}],"datasets":[{"url":"/dataset/cityscapes","name":"Cityscapes","full_name":"","num_papers_in_archive":3702},{"url":"/dataset/nuscenes","name":"nuScenes","full_name":"","num_papers_in_archive":2139},{"url":"/dataset/ade20k","name":"ADE20K","full_name":"","num_papers_in_archive":1213},{"url":"/dataset/semantickitti","name":"SemanticKITTI","full_name":"","num_papers_in_archive":669},{"url":"/dataset/2d-3d-s","name":"2D-3D-S","full_name":"2D-3D-Semantic","num_papers_in_archive":147},{"url":"/dataset/woodscape","name":"WoodScape","full_name":"","num_papers_in_archive":55},{"url":"/dataset/imagenet-s","name":"ImageNet-S","full_name":"ImageNet Semantic Segmentation","num_papers_in_archive":43},{"url":"/dataset/suim","name":"SUIM","full_name":"Segmentation of Underwater IMagery","num_papers_in_archive":34},{"url":"/dataset/2017-robotic-instrument-segmentation","name":"2017 Robotic Instrument Segmentation Challenge","full_name":"","num_papers_in_archive":22},{"url":"/dataset/kvasir-instrument","name":"Kvasir-Instrument","full_name":"","num_papers_in_archive":19},{"url":"/dataset/scribblekitti","name":"ScribbleKITTI","full_name":"","num_papers_in_archive":19},{"url":"/dataset/kits19","name":"KiTS19","full_name":"The 2019 Kidney and Kidney Tumor Segmentation Challenge","num_papers_in_archive":7},{"url":"/dataset/dronescapes","name":"Dronescapes","full_name":"","num_papers_in_archive":0}],"subtasks":[],"parent_tasks":[{"url":"/task/semantic-segmentation","name":"Semantic Segmentation"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":109,"tagged_in_all":190,"items":[{"url":"/paper/adversarial-learning-for-semi-supervised","title":"Adversarial Learning for Semi-Supervised Semantic Segmentation","date":"2018-02-22","arxiv_id":"1802.07934","repositories_listed":13,"syntology":{"n":1,"n_ran":1,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/mean-teachers-are-better-role-models-weight","title":"Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results","date":"2017-03-06","arxiv_id":"1703.01780","repositories_listed":8,"syntology":{"n":6,"n_ran":6,"n_unverified":0,"n_pointer_only":6}},{"url":"/paper/semi-supervised-semantic-segmentation-with-1","title":"Semi-Supervised Semantic Segmentation with Cross-Consistency Training","date":"2020-03-19","arxiv_id":"2003.09005","repositories_listed":5,"syntology":{"n":20,"n_ran":10,"n_unverified":10,"n_pointer_only":0}},{"url":"/paper/consistency-regularization-and-cutmix-for","title":"Semi-supervised semantic segmentation needs strong, varied perturbations","date":"2019-06-05","arxiv_id":"1906.01916","repositories_listed":5,"syntology":{"n":3,"n_ran":0,"n_unverified":3,"n_pointer_only":0}},{"url":"/paper/semi-supervised-semantic-segmentation-with-3","title":"Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision","date":"2021-06-02","arxiv_id":"2106.01226","repositories_listed":3,"syntology":{"n":39,"n_ran":22,"n_unverified":17,"n_pointer_only":28}},{"url":"/paper/fast-online-object-tracking-and-segmentation","title":"Fast Online Object Tracking and Segmentation: A Unifying Approach","date":"2018-12-12","arxiv_id":"1812.05050","repositories_listed":3,"syntology":{"n":10,"n_ran":1,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/decoupled-deep-neural-network-for-semi","title":"Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation","date":"2015-06-16","arxiv_id":"1506.04924","repositories_listed":3,"syntology":null},{"url":"/paper/weakly-and-semi-supervised-learning-of-a-dcnn","title":"Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation","date":"2015-02-09","arxiv_id":"1502.02734","repositories_listed":3,"syntology":null},{"url":"/paper/beyond-pixels-semi-supervised-semantic","title":"Beyond Pixels: Semi-Supervised Semantic Segmentation with a Multi-scale Patch-based Multi-Label Classifier","date":"2024-07-04","arxiv_id":"2407.04036","repositories_listed":2,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/bidirectional-copy-paste-for-semi-supervised","title":"Bidirectional Copy-Paste for Semi-Supervised Medical Image Segmentation","date":"2023-05-01","arxiv_id":"2305.00673","repositories_listed":2,"syntology":{"n":19,"n_ran":6,"n_unverified":13,"n_pointer_only":0}},{"url":"/paper/conflict-based-cross-view-consistency-for","title":"Conflict-Based Cross-View Consistency for Semi-Supervised Semantic Segmentation","date":"2023-03-02","arxiv_id":"2303.01276","repositories_listed":2,"syntology":{"n":2,"n_ran":0,"n_unverified":2,"n_pointer_only":2}},{"url":"/paper/pseudo-label-guided-contrastive-learning-for","title":"Pseudo-Label Guided Contrastive Learning for Semi-Supervised Medical Image Segmentation","date":"2023-01-01","arxiv_id":null,"repositories_listed":2,"syntology":null},{"url":"/paper/triple-view-feature-learning-for-medical","title":"Triple-View Feature Learning for Medical Image Segmentation","date":"2022-08-12","arxiv_id":"2208.06303","repositories_listed":2,"syntology":null},{"url":"/paper/lasermix-for-semi-supervised-lidar-semantic","title":"LaserMix for Semi-Supervised LiDAR Semantic Segmentation","date":"2022-06-30","arxiv_id":"2207.00026","repositories_listed":2,"syntology":null},{"url":"/paper/translation-consistent-semi-supervised","title":"Translation Consistent Semi-supervised Segmentation for 3D Medical Images","date":"2022-03-28","arxiv_id":"2203.14523","repositories_listed":2,"syntology":{"n":9,"n_ran":0,"n_unverified":9,"n_pointer_only":0}},{"url":"/paper/reliable-semantic-segmentation-with","title":"Robust Semantic Segmentation with Superpixel-Mix","date":"2021-08-02","arxiv_id":"2108.00968","repositories_listed":2,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}},{"url":"/paper/semi-supervised-semantic-segmentation-with-4","title":"Semi-supervised Semantic Segmentation with Directional Context-aware Consistency","date":"2021-06-27","arxiv_id":"2106.14133","repositories_listed":2,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/bootstrapping-semantic-segmentation-with","title":"Bootstrapping Semantic Segmentation with Regional Contrast","date":"2021-04-09","arxiv_id":"2104.04465","repositories_listed":2,"syntology":{"n":6,"n_ran":3,"n_unverified":3,"n_pointer_only":2}},{"url":"/paper/cylindrical-and-asymmetrical-3d-convolution","title":"Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation","date":"2020-11-19","arxiv_id":"2011.10033","repositories_listed":2,"syntology":null},{"url":"/paper/pseudoseg-designing-pseudo-labels-for-1","title":"PseudoSeg: Designing Pseudo Labels for Semantic Segmentation","date":"2020-10-19","arxiv_id":"2010.09713","repositories_listed":2,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/classmix-segmentation-based-data-augmentation","title":"ClassMix: Segmentation-Based Data Augmentation for Semi-Supervised Learning","date":"2020-07-15","arxiv_id":"2007.07936","repositories_listed":2,"syntology":{"n":19,"n_ran":5,"n_unverified":14,"n_pointer_only":0}},{"url":"/paper/part-aware-prototype-network-for-few-shot","title":"Part-aware Prototype Network for Few-shot Semantic Segmentation","date":"2020-07-13","arxiv_id":"2007.06309","repositories_listed":2,"syntology":null},{"url":"/paper/dearli-decoupled-enhancement-of-recognition","title":"DEARLi: Decoupled Enhancement of Recognition and Localization for Semi-supervised Panoptic Segmentation","date":"2025-07-14","arxiv_id":"2507.10118","repositories_listed":1,"syntology":null},{"url":"/paper/leveraging-out-of-distribution-unlabeled","title":"Leveraging Out-of-Distribution Unlabeled Images: Semi-Supervised Semantic Segmentation with an Open-Vocabulary Model","date":"2025-07-04","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/farcluss-fuzzy-adaptive-rebalancing-and","title":"FARCLUSS: Fuzzy Adaptive Rebalancing and Contrastive Uncertainty Learning for Semi-Supervised Semantic Segmentation","date":"2025-06-11","arxiv_id":"2506.11142","repositories_listed":1,"syntology":null},{"url":"/paper/rs-mtdf-multi-teacher-distillation-and-fusion","title":"RS-MTDF: Multi-Teacher Distillation and Fusion for Remote Sensing Semi-Supervised Semantic Segmentation","date":"2025-06-10","arxiv_id":"2506.08772","repositories_listed":1,"syntology":null},{"url":"/paper/semi-supervised-semantic-segmentation-with-11","title":"Semi-supervised Semantic Segmentation with Multi-Constraint Consistency Learning","date":"2025-03-23","arxiv_id":"2503.17914","repositories_listed":1,"syntology":null},{"url":"/paper/exploring-token-level-augmentation-in-vision","title":"Exploring Token-Level Augmentation in Vision Transformer for Semi-Supervised Semantic Segmentation","date":"2025-03-04","arxiv_id":"2503.02459","repositories_listed":1,"syntology":null},{"url":"/paper/confidence-weighted-boundary-aware-learning","title":"Confidence-Weighted Boundary-Aware Learning for Semi-Supervised Semantic Segmentation","date":"2025-02-21","arxiv_id":"2502.15152","repositories_listed":1,"syntology":null},{"url":"/paper/semi-supervised-semantic-segmentation-for","title":"Semi-supervised Semantic Segmentation for Remote Sensing Images via Multi-scale Uncertainty Consistency and Cross-Teacher-Student Attention","date":"2025-01-18","arxiv_id":"2501.10736","repositories_listed":1,"syntology":null}],"syntology_records":15,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}