{"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/co-occurrence-feature-learning-from-skeleton","title":"Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation","arxiv_id":"1804.06055","date":"2018-04-17","proceeding":null,"authors":["Chao Li","Qiaoyong Zhong","Di Xie","ShiLiang Pu"],"abstract":"Skeleton-based human action recognition has recently drawn increasing\nattentions with the availability of large-scale skeleton datasets. The most\ncrucial factors for this task lie in two aspects: the intra-frame\nrepresentation for joint co-occurrences and the inter-frame representation for\nskeletons' temporal evolutions. In this paper we propose an end-to-end\nconvolutional co-occurrence feature learning framework. The co-occurrence\nfeatures are learned with a hierarchical methodology, in which different levels\nof contextual information are aggregated gradually. Firstly point-level\ninformation of each joint is encoded independently. Then they are assembled\ninto semantic representation in both spatial and temporal domains.\nSpecifically, we introduce a global spatial aggregation scheme, which is able\nto learn superior joint co-occurrence features over local aggregation. Besides,\nraw skeleton coordinates as well as their temporal difference are integrated\nwith a two-stream paradigm. Experiments show that our approach consistently\noutperforms other state-of-the-arts on action recognition and detection\nbenchmarks like NTU RGB+D, SBU Kinect Interaction and PKU-MMD.","url_abs":"http://arxiv.org/abs/1804.06055v1","url_pdf":"http://arxiv.org/pdf/1804.06055v1.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":"co-occurrence-feature-learning-from-skeleton","repo_url":"https://github.com/hikvision-research/skelact","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"co-occurrence-feature-learning-from-skeleton","repo_url":"https://github.com/fandulu/Keras-for-Co-occurrence-Feature-Learning-from-Skeleton-Data-for-Action-Recognition","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"co-occurrence-feature-learning-from-skeleton","repo_url":"https://github.com/hhe-distance/AIF-CNN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"co-occurrence-feature-learning-from-skeleton","repo_url":"https://github.com/huguyuehuhu/HCN-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"co-occurrence-feature-learning-from-skeleton","repo_url":"https://github.com/maxstrobel/HCN-PrototypeLoss-PyTorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"co-occurrence-feature-learning-from-skeleton","repo_url":"https://github.com/natepuppy/HCN-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"rf-based-pose-estimation","task_name":"RF-based Pose Estimation"},{"task_slug":"skeleton-based-action-recognition","task_name":"Skeleton Based Action Recognition"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/rf-based-pose-estimation-on-rf-mmd","task":"RF-based Pose Estimation","dataset":"RF-MMD","model":"HCN","rank_in_archive_order":1,"of":2,"metrics":{"mAP (@0.1, Through-wall)":"78.5","mAP (@0.1, Visible)":"82,5"},"uses_additional_data":false},{"leaderboard":"/sota/skeleton-based-action-recognition-on-ntu-rgbd","task":"Skeleton Based Action Recognition","dataset":"NTU RGB+D","model":"HCN","rank_in_archive_order":84,"of":135,"metrics":{"Accuracy (CS)":"86.5","Accuracy (CV)":"91.1"},"uses_additional_data":false},{"leaderboard":"/sota/skeleton-based-action-recognition-on-pku-mmd","task":"Skeleton Based Action Recognition","dataset":"PKU-MMD","model":"HCN","rank_in_archive_order":2,"of":4,"metrics":{"mAP@0.50 (CS)":"92.6","mAP@0.50 (CV)":"94.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.06055","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1804.06055"}},"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/huguyuehuhu/HCN-pytorch","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/natepuppy/HCN-pytorch","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/fandulu/Keras-for-Co-occurrence-Feature-Learning-from-Skeleton-Data-for-Action-Recognition","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/maxstrobel/HCN-PrototypeLoss-PyTorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hikvision-research/skelact","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/hhe-distance/AIF-CNN","reach":null}],"summary":{"ran_fixture":1,"ran":1,"ran_draft_wrong":1,"ran_honours":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":2,"repositories":1},"listed":{"samples":3,"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":3,"samples":[{"code_sha256_prefix":"9850c4e85f05e9be","entry":"accuracytop1","repo":"hhe-distance/AIF-CNN","repo_kind":"listed","path":"for_ntu/model/AIF_CNN.py","file_url":"https://github.com/hhe-distance/AIF-CNN/blob/HEAD/for_ntu/model/AIF_CNN.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9850c4e85f05e9be"}},{"code_sha256_prefix":"a705f16bf740c3fb","entry":"identity","repo":"hikvision-research/skelact","repo_kind":"official","path":"skelact/models/backbones/dgcn.py","file_url":"https://github.com/hikvision-research/skelact/blob/HEAD/skelact/models/backbones/dgcn.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a705f16bf740c3fb"}},{"code_sha256_prefix":"420f3c8dfea9f457","entry":"loss_fn","repo":"hhe-distance/AIF-CNN","repo_kind":"listed","path":"for_ntu/model/AIF_CNN.py","file_url":"https://github.com/hhe-distance/AIF-CNN/blob/HEAD/for_ntu/model/AIF_CNN.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"420f3c8dfea9f457"}},{"code_sha256_prefix":"cf1326f21a3b15f8","entry":"zero","repo":"hikvision-research/skelact","repo_kind":"official","path":"skelact/models/backbones/dgcn.py","file_url":"https://github.com/hikvision-research/skelact/blob/HEAD/skelact/models/backbones/dgcn.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cf1326f21a3b15f8"}},{"code_sha256_prefix":"c43500bf04556120","entry":"accuracytop2","repo":"hhe-distance/AIF-CNN","repo_kind":"listed","path":"for_ntu/model/AIF_CNN.py","file_url":"https://github.com/hhe-distance/AIF-CNN/blob/HEAD/for_ntu/model/AIF_CNN.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c43500bf04556120"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}