{"url":"/sota/human-instance-segmentation-on-ochuman","task":{"name":"Human Instance Segmentation","url":"/task/human-instance-segmentation","note":null},"dataset":{"name":"OCHuman","url":"/dataset/ochuman"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Instance segmentation is the task of detecting and delineating each distinct object of interest appearing in an image.\r\n\r\nImage Credit: [Deep Occlusion-Aware Instance Segmentation with Overlapping BiLayers](https://arxiv.org/abs/2103.12340)","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":["AP"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"AP":"higher"}},"counts":{"rows":18,"rows_with_code":10,"rows_with_paper_page":18,"rows_dated":16,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"BBox-Mask-Pose 2x","metrics":{"AP":"32.4"},"uses_additional_data":false,"paper_date":"2024-12-02","paper":"/paper/detection-pose-estimation-and-segmentation-1","paper_url":"https://arxiv.org/abs/2412.01562v1","paper_title":"Detection, Pose Estimation and Segmentation for Multiple Bodies: Closing the Virtuous Circle","code":"https://github.com/MiraPurkrabek/BBoxMaskPose","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"Crowd-SAM (ViT-L)","metrics":{"AP":"31.4"},"uses_additional_data":false,"paper_date":"2024-07-16","paper":"/paper/crowd-sam-sam-as-a-smart-annotator-for-object","paper_url":"https://arxiv.org/abs/2407.11464v2","paper_title":"Crowd-SAM: SAM as a Smart Annotator for Object Detection in Crowded Scenes","code":"https://github.com/felixcaae/crowdsam","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"HQNet (ResNet-50)","metrics":{"AP":"31.1"},"uses_additional_data":false,"paper_date":"2023-12-09","paper":"/paper/you-only-learn-one-query-learning-unified","paper_url":"https://arxiv.org/abs/2312.05525v3","paper_title":"You Only Learn One Query: Learning Unified Human Query for Single-Stage Multi-Person Multi-Task Human-Centric Perception","code":"https://github.com/lishuhuai527/coco-unihuman","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"BlendMask + CIS","metrics":{"AP":"29.8"},"uses_additional_data":false,"paper_date":"2021-08-16","paper":"/paper/real-time-human-centric-segmentation-for","paper_url":"https://arxiv.org/abs/2108.07199v1","paper_title":"Real-time Human-Centric Segmentation for Complex Video Scenes","code":"https://github.com/iigroup/hvisnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"Mask2Former + Occlusion C&P","metrics":{"AP":"28.3"},"uses_additional_data":false,"paper_date":"2022-10-07","paper":"/paper/humans-need-not-label-more-humans-occlusion","paper_url":"https://arxiv.org/abs/2210.03686v1","paper_title":"Humans need not label more humans: Occlusion Copy & Paste for Occluded Human Instance Segmentation","code":"https://github.com/levan92/occlusion-copy-paste","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":4,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"CondInst + CIS","metrics":{"AP":"28.1"},"uses_additional_data":false,"paper_date":"2021-08-16","paper":"/paper/real-time-human-centric-segmentation-for","paper_url":"https://arxiv.org/abs/2108.07199v1","paper_title":"Real-time Human-Centric Segmentation for Complex Video Scenes","code":"https://github.com/iigroup/hvisnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":7,"model":"Mask2Former","metrics":{"AP":"27.8"},"uses_additional_data":false,"paper_date":"2023-03-13","paper":"/paper/object-centric-multi-task-learning-for-human","paper_url":"https://arxiv.org/abs/2303.06800v1","paper_title":"Object-Centric Multi-Task Learning for Human Instances","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"HCQNet","metrics":{"AP":"27.3"},"uses_additional_data":false,"paper_date":"2023-03-13","paper":"/paper/object-centric-multi-task-learning-for-human","paper_url":"https://arxiv.org/abs/2303.06800v1","paper_title":"Object-Centric Multi-Task Learning for Human Instances","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"ExPoSeg","metrics":{"AP":"26.8"},"uses_additional_data":false,"paper_date":"2020-01-07","paper":"/paper/poseg-pose-aware-refinement-network-for-human","paper_url":"https://ieeexplore.ieee.org/document/8962018","paper_title":"PoSeg: Pose-Aware Refinement Network for Human Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":10,"model":"RTMDet-ins-l","metrics":{"AP":"26.5"},"uses_additional_data":false,"paper_date":"2024-12-02","paper":"/paper/detection-pose-estimation-and-segmentation-1","paper_url":"https://arxiv.org/abs/2412.01562v1","paper_title":"Detection, Pose Estimation and Segmentation for Multiple Bodies: Closing the Virtuous Circle","code":"https://github.com/MiraPurkrabek/BBoxMaskPose","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"JoPoSeg","metrics":{"AP":"26.4"},"uses_additional_data":false,"paper_date":"2020-01-07","paper":"/paper/poseg-pose-aware-refinement-network-for-human","paper_url":"https://ieeexplore.ieee.org/document/8962018","paper_title":"PoSeg: Pose-Aware Refinement Network for Human Instance Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"BaseNet-DPS","metrics":{"AP":"25.5"},"uses_additional_data":false,"paper_date":"2023-03-13","paper":"/paper/object-centric-multi-task-learning-for-human","paper_url":"https://arxiv.org/abs/2303.06800v1","paper_title":"Object-Centric Multi-Task Learning for Human Instances","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"Pose2Seg","metrics":{"AP":"23.8"},"uses_additional_data":false,"paper_date":"2018-03-28","paper":"/paper/pose2seg-detection-free-human-instance","paper_url":"http://arxiv.org/abs/1803.10683v3","paper_title":"Pose2Seg: Detection Free Human Instance Segmentation","code":"https://github.com/open-mmlab/mmpose","n_code_links":7,"syntology":{"n_ran":0,"n_unverified":16,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"PolarMask + CIS","metrics":{"AP":"23.4"},"uses_additional_data":false,"paper_date":"2021-08-16","paper":"/paper/real-time-human-centric-segmentation-for","paper_url":"https://arxiv.org/abs/2108.07199v1","paper_title":"Real-time Human-Centric Segmentation for Complex Video Scenes","code":"https://github.com/iigroup/hvisnet","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"ResNet-101-FPN + TTG v1","metrics":{"AP":"22.42"},"uses_additional_data":false,"paper_date":"2022-12-12","paper":"/paper/test-time-adaptation-vs-training-time","paper_url":"https://arxiv.org/abs/2212.06242v1","paper_title":"Test-time Adaptation vs. Training-time Generalization: A Case Study in Human Instance Segmentation using Keypoints Estimation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":16,"model":"BCNet","metrics":{"AP":"20.6"},"uses_additional_data":false,"paper_date":"2022-08-08","paper":"/paper/occlusion-aware-instance-segmentation-via","paper_url":"https://arxiv.org/abs/2208.04438v2","paper_title":"Occlusion-Aware Instance Segmentation via BiLayer Network Architectures","code":"https://github.com/lkeab/BCNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"CaSe","metrics":{"AP":"18.0"},"uses_additional_data":false,"paper_date":null,"paper":"/paper/count-and-similarity-aware-r-cnn-for","paper_url":"https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2678_ECCV_2020_paper.php","paper_title":"Count- and Similarity-aware R-CNN for Pedestrian Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"Mask RCNN","metrics":{"AP":"16.9"},"uses_additional_data":false,"paper_date":null,"paper":"/paper/count-and-similarity-aware-r-cnn-for","paper_url":"https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/2678_ECCV_2020_paper.php","paper_title":"Count- and Similarity-aware R-CNN for Pedestrian Detection","code":null,"n_code_links":0,"syntology":null}],"since_archive":{"present":false,"note":"No Syntology-extracted rows are published in this build."},"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":2,"rows_with_any_sample_ran":0,"distinct_papers_with_graph_line":2,"distinct_papers_with_any_sample_ran":0,"samples_over_distinct_papers":{"n_ran":0,"n_unverified":20,"n_samples":20,"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":0,"n_unverified":20,"n_samples":20,"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"}}}