{"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/pose2seg-detection-free-human-instance","title":"Pose2Seg: Detection Free Human Instance Segmentation","arxiv_id":"1803.10683","date":"2018-03-28","proceeding":"CVPR 2019 6","authors":["Song-Hai Zhang","Rui-Long Li","Xin Dong","Paul L. Rosin","Zixi Cai","Han Xi","Dingcheng Yang","Hao-Zhi Huang","Shi-Min Hu"],"abstract":"The standard approach to image instance segmentation is to perform the object\ndetection first, and then segment the object from the detection bounding-box.\nMore recently, deep learning methods like Mask R-CNN perform them jointly.\nHowever, little research takes into account the uniqueness of the \"human\"\ncategory, which can be well defined by the pose skeleton. Moreover, the human\npose skeleton can be used to better distinguish instances with heavy occlusion\nthan using bounding-boxes. In this paper, we present a brand new pose-based\ninstance segmentation framework for humans which separates instances based on\nhuman pose, rather than proposal region detection. We demonstrate that our\npose-based framework can achieve better accuracy than the state-of-art\ndetection-based approach on the human instance segmentation problem, and can\nmoreover better handle occlusion. Furthermore, there are few public datasets\ncontaining many heavily occluded humans along with comprehensive annotations,\nwhich makes this a challenging problem seldom noticed by researchers.\nTherefore, in this paper we introduce a new benchmark \"Occluded Human\n(OCHuman)\", which focuses on occluded humans with comprehensive annotations\nincluding bounding-box, human pose and instance masks. This dataset contains\n8110 detailed annotated human instances within 4731 images. With an average\n0.67 MaxIoU for each person, OCHuman is the most complex and challenging\ndataset related to human instance segmentation. Through this dataset, we want\nto emphasize occlusion as a challenging problem for researchers to study.","url_abs":"http://arxiv.org/abs/1803.10683v3","url_pdf":"http://arxiv.org/pdf/1803.10683v3.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":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/liruilong940607/OCHumanApi","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/liruilong940607/Pose2Seg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/Jittor/InstanceSegmentation-jittor","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/hz-ants/Pose2Seg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/jacksonlli/ProjectQuarantine_HumanSegmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/ligaoqi2/Pose2Seg-single-person-video-demo","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pose2seg-detection-free-human-instance","repo_url":"https://github.com/open-mmlab/mmpose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"2d-human-pose-estimation","task_name":"2D Human Pose Estimation"},{"task_slug":"human-instance-segmentation","task_name":"Human Instance Segmentation"},{"task_slug":"instance-segmentation","task_name":"Instance Segmentation"},{"task_slug":"keypoint-detection","task_name":"Keypoint Detection"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"pose-based-human-instance-segmentation","task_name":"Pose-Based Human Instance Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"roi-align","method_name":"RoIAlign"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[{"slug":"ochuman","name":"OCHuman","full_name":null}],"methods_introduced":[],"results":[{"leaderboard":"/sota/2d-human-pose-estimation-on-ochuman","task":"2D Human Pose Estimation","dataset":"OCHuman","model":"Pose2Seg","rank_in_archive_order":11,"of":11,"metrics":{"Test AP":"23.8"},"uses_additional_data":false},{"leaderboard":"/sota/human-instance-segmentation-on-ochuman","task":"Human Instance Segmentation","dataset":"OCHuman","model":"Pose2Seg","rank_in_archive_order":13,"of":18,"metrics":{"AP":"23.8"},"uses_additional_data":false},{"leaderboard":"/sota/keypoint-detection-on-ochuman","task":"Keypoint Detection","dataset":"OCHuman","model":"Pose2Seg","rank_in_archive_order":10,"of":10,"metrics":{"Test AP":"23.8"},"uses_additional_data":false},{"leaderboard":"/sota/pose-based-human-instance-segmentation-on","task":"Pose-Based Human Instance Segmentation","dataset":"OCHuman","model":"Pose2Seg (plus ground-truth keypoints)","rank_in_archive_order":1,"of":1,"metrics":{"AP":"55.2"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1803.10683","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1803.10683"}},"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/open-mmlab/mmpose","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ligaoqi2/Pose2Seg-single-person-video-demo","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/liruilong940607/OCHumanApi","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Jittor/InstanceSegmentation-jittor","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hz-ants/Pose2Seg","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jacksonlli/ProjectQuarantine_HumanSegmentation","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/liruilong940607/Pose2Seg","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":16},"by_repo_kind":{"official":{"samples":16,"ran":0,"repositories":2}},"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":"5f9acf5c73a56bba","entry":"adjust_learning_rate","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/torch_utils.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/torch_utils.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":"5f9acf5c73a56bba"}},{"code_sha256_prefix":"3adc733362acbc3a","entry":"affine_align_gpu","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"modeling/affine_align.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/modeling/affine_align.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":"3adc733362acbc3a"}},{"code_sha256_prefix":"b4424a7aa52699a1","entry":"cluster_zixi","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"cluster_pose.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/cluster_pose.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":"b4424a7aa52699a1"}},{"code_sha256_prefix":"7ea3f744b9047938","entry":"draw_bbox","repo":"liruilong940607/OCHumanApi","repo_kind":"official","path":"ochumanApi/vis.py","file_url":"https://github.com/liruilong940607/OCHumanApi/blob/HEAD/ochumanApi/vis.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":"7ea3f744b9047938"}},{"code_sha256_prefix":"ac93919e94bebaf3","entry":"draw_mask","repo":"liruilong940607/OCHumanApi","repo_kind":"official","path":"ochumanApi/vis.py","file_url":"https://github.com/liruilong940607/OCHumanApi/blob/HEAD/ochumanApi/vis.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":"ac93919e94bebaf3"}},{"code_sha256_prefix":"0270566392d20661","entry":"draw_skeleton","repo":"liruilong940607/OCHumanApi","repo_kind":"official","path":"ochumanApi/vis.py","file_url":"https://github.com/liruilong940607/OCHumanApi/blob/HEAD/ochumanApi/vis.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":"0270566392d20661"}},{"code_sha256_prefix":"28182a3c822c54b8","entry":"draw_skeleton","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"cluster_pose.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/cluster_pose.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":"28182a3c822c54b8"}},{"code_sha256_prefix":"97b06bbd92908d15","entry":"get_affine_matrix","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/transforms.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/transforms.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":"97b06bbd92908d15"}},{"code_sha256_prefix":"929bb26e108c70d6","entry":"get_aug_matrix","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/transforms.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/transforms.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":"929bb26e108c70d6"}},{"code_sha256_prefix":"cb93dd0e2cf85645","entry":"masks2bboxes","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/np_utils.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/np_utils.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":"cb93dd0e2cf85645"}},{"code_sha256_prefix":"06275974c5bef2dc","entry":"pad","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/np_utils.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/np_utils.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":"06275974c5bef2dc"}},{"code_sha256_prefix":"856af44302c4de93","entry":"resize_keep_ratio","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/np_utils.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/np_utils.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":"856af44302c4de93"}},{"code_sha256_prefix":"ca039bc7733a9c29","entry":"solve_affinematrix","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"modeling/core.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/modeling/core.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":"ca039bc7733a9c29"}},{"code_sha256_prefix":"b0526d917e08136a","entry":"to_np","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/torch_utils.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/torch_utils.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":"b0526d917e08136a"}},{"code_sha256_prefix":"f62d83b95f3740c6","entry":"to_var","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/torch_utils.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/torch_utils.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":"f62d83b95f3740c6"}},{"code_sha256_prefix":"e5661f2323631e7b","entry":"warpAffinePoints","repo":"liruilong940607/Pose2Seg","repo_kind":"official","path":"lib/transforms.py","file_url":"https://github.com/liruilong940607/Pose2Seg/blob/HEAD/lib/transforms.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":"e5661f2323631e7b"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}