{"url":"/sota/image-level-supervised-instance-segmentation","task":{"name":"Image-level Supervised Instance Segmentation","url":"/task/image-level-supervised-instance-segmentation","note":null},"dataset":{"name":"PASCAL VOC 2012 val","url":null},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Weakly-Supervised Instance Segmentation using Image-level Labels","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["mAP@0.5","mAP@0.25","mAP@0.7","mAP@0.75"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mAP@0.5":"higher","mAP@0.25":"higher","mAP@0.7":"higher","mAP@0.75":"higher"}},"counts":{"rows":13,"rows_with_code":9,"rows_with_paper_page":13,"rows_dated":13,"rows_using_additional_data":9},"rows":[{"rank_in_archive_order":1,"model":"WeakSAM-Mask2Former (with SAM)","metrics":{"mAP@0.25":"73.4","mAP@0.5":"64.4","mAP@0.7":"49.7","mAP@0.75":"45.3"},"uses_additional_data":false,"paper_date":"2024-02-22","paper":"/paper/weaksam-segment-anything-meets-weakly","paper_url":"https://arxiv.org/abs/2402.14812v2","paper_title":"WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition","code":"https://github.com/hustvl/weaksam","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"WeakSAM-Mask RCNN (with SAM)","metrics":{"mAP@0.25":"70.3","mAP@0.5":"59.6","mAP@0.7":"43.1","mAP@0.75":"36.2"},"uses_additional_data":false,"paper_date":"2024-02-22","paper":"/paper/weaksam-segment-anything-meets-weakly","paper_url":"https://arxiv.org/abs/2402.14812v2","paper_title":"WeakSAM: Segment Anything Meets Weakly-supervised Instance-level Recognition","code":"https://github.com/hustvl/weaksam","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"CIM + Mask R-CNN","metrics":{"mAP@0.25":"68.7","mAP@0.5":"55.9","mAP@0.7":"37.1","mAP@0.75":"30.9"},"uses_additional_data":true,"paper_date":"2024-02-12","paper":"/paper/complete-instances-mining-for-weakly-1","paper_url":"https://arxiv.org/abs/2402.07633v1","paper_title":"Complete Instances Mining for Weakly Supervised Instance Segmentation","code":"https://github.com/ZechengLi19/CIM","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":1,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"CIM","metrics":{"mAP@0.25":"64.9","mAP@0.5":"51.1","mAP@0.7":"32.4","mAP@0.75":"26.1"},"uses_additional_data":true,"paper_date":"2024-02-12","paper":"/paper/complete-instances-mining-for-weakly-1","paper_url":"https://arxiv.org/abs/2402.07633v1","paper_title":"Complete Instances Mining for Weakly Supervised Instance Segmentation","code":"https://github.com/ZechengLi19/CIM","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":1,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"BESTIE (proposal-free)","metrics":{"mAP@0.25":"61.2","mAP@0.5":"51.0","mAP@0.7":"31.9","mAP@0.75":"26.6"},"uses_additional_data":false,"paper_date":"2021-09-20","paper":"/paper/beyond-semantic-to-instance-segmentation","paper_url":"https://arxiv.org/abs/2109.09477v3","paper_title":"Beyond Semantic to Instance Segmentation: Weakly-Supervised Instance Segmentation via Semantic Knowledge Transfer and Self-Refinement","code":"https://github.com/clovaai/BESTIE","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"Arun et al.","metrics":{"mAP@0.25":"59.7","mAP@0.5":"50.9","mAP@0.7":"30.2","mAP@0.75":"28.5"},"uses_additional_data":true,"paper_date":"2020-07-18","paper":"/paper/weakly-supervised-instance-segmentation-by-1","paper_url":"https://arxiv.org/abs/2007.09397v1","paper_title":"Weakly Supervised Instance Segmentation by Learning Annotation Consistent Instances","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"LIID","metrics":{"mAP@0.5":"48.4","mAP@0.75":"24.9"},"uses_additional_data":true,"paper_date":"2020-09-10","paper":"/paper/leveraging-instance-image-and-dataset-level","paper_url":"https://ieeexplore.ieee.org/abstract/document/9193980","paper_title":"Leveraging Instance-, Image- and Dataset-Level Information for Weakly Supervised Instance Segmentation","code":"https://github.com/yun-liu/LIID","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"IRN (proposal-free)","metrics":{"mAP@0.5":"46.7","mAP@0.7":"23.5"},"uses_additional_data":false,"paper_date":"2019-04-10","paper":"/paper/weakly-supervised-learning-of-instance","paper_url":"https://arxiv.org/abs/1904.05044v3","paper_title":"Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations","code":"https://github.com/jiwoon-ahn/irn","n_code_links":8,"syntology":{"n_ran":1,"n_unverified":9,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"WISE","metrics":{"mAP@0.25":"49.2","mAP@0.5":"41.7","mAP@0.75":"23.7"},"uses_additional_data":true,"paper_date":"2019-07-02","paper":"/paper/where-are-the-masks-instance-segmentation","paper_url":"https://arxiv.org/abs/1907.01430v1","paper_title":"Where are the Masks: Instance Segmentation with Image-level Supervision","code":"https://github.com/ElementAI/wise_ils","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"CL","metrics":{"mAP@0.25":"56.6","mAP@0.5":"38.1","mAP@0.75":"12.3"},"uses_additional_data":true,"paper_date":"2020-01-30","paper":"/paper/weakly-supervised-instance-segmentation-by","paper_url":"https://arxiv.org/abs/2001.11207v3","paper_title":"Weakly Supervised Instance Segmentation by Deep Community Learning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":11,"model":"Label-PEnet","metrics":{"mAP@0.25":"49.2","mAP@0.5":"30.2","mAP@0.75":"12.9"},"uses_additional_data":true,"paper_date":"2019-10-07","paper":"/paper/label-penet-sequential-label-propagation-and","paper_url":"https://arxiv.org/abs/1910.02624v3","paper_title":"Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"IAM","metrics":{"mAP@0.25":"45.9","mAP@0.5":"28.3","mAP@0.75":"11.9"},"uses_additional_data":true,"paper_date":"2019-06-01","paper":"/paper/learning-instance-activation-maps-for-weakly","paper_url":"http://openaccess.thecvf.com/content_CVPR_2019/html/Zhu_Learning_Instance_Activation_Maps_for_Weakly_Supervised_Instance_Segmentation_CVPR_2019_paper.html","paper_title":"Learning Instance Activation Maps for Weakly Supervised Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"PRM","metrics":{"mAP@0.25":"44.3","mAP@0.5":"26.8","mAP@0.75":"9.0"},"uses_additional_data":true,"paper_date":"2018-04-03","paper":"/paper/weakly-supervised-instance-segmentation-using-1","paper_url":"http://arxiv.org/abs/1804.00880v1","paper_title":"Weakly Supervised Instance Segmentation using Class Peak Response","code":"https://github.com/ZhouYanzhao/PRM","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":3,"rows_with_any_sample_ran":3,"distinct_papers_with_graph_line":2,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":9,"n_unverified":10,"n_samples":19,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":17,"n_unverified":11,"n_samples":28,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}