{"url":"/sota/online-action-detection-on-thumos-14","task":{"name":"Online Action Detection","url":"/task/online-action-detection","note":null},"dataset":{"name":"THUMOS'14","url":"/dataset/thumos14-1"},"category":"Computer Vision","categories":["Computer Vision","Time Series"],"category_note":null,"description":"Online action detection is the task of predicting the action as soon as it happens in a streaming video without access to video frames in the future.","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","MFLOPs per pred"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"mAP":"higher","MFLOPs per pred":null}},"counts":{"rows":15,"rows_with_code":11,"rows_with_paper_page":15,"rows_dated":15,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"GateHUB(TimeSformer)","metrics":{"mAP":"72.5"},"uses_additional_data":false,"paper_date":"2022-06-09","paper":"/paper/gatehub-gated-history-unit-with-background-1","paper_url":"https://arxiv.org/abs/2206.04668v1","paper_title":"GateHUB: Gated History Unit with Background Suppression for Online Action Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"MiniROAD","metrics":{"MFLOPs per pred":"15.8","mAP":"71.8"},"uses_additional_data":false,"paper_date":"2023-01-01","paper":"/paper/miniroad-minimal-rnn-framework-for-online","paper_url":"http://openaccess.thecvf.com//content/ICCV2023/html/An_MiniROAD_Minimal_RNN_Framework_for_Online_Action_Detection_ICCV_2023_paper.html","paper_title":"MiniROAD: Minimal RNN Framework for Online Action Detection","code":"https://github.com/jbistanbul/miniroad","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"GateHUB","metrics":{"mAP":"70.7"},"uses_additional_data":false,"paper_date":"2022-06-09","paper":"/paper/gatehub-gated-history-unit-with-background-1","paper_url":"https://arxiv.org/abs/2206.04668v1","paper_title":"GateHUB: Gated History Unit with Background Suppression for Online Action Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"LSTR","metrics":{"mAP":"69.5"},"uses_additional_data":false,"paper_date":"2021-07-07","paper":"/paper/long-short-term-transformer-for-online-action","paper_url":"https://arxiv.org/abs/2107.03377v3","paper_title":"Long Short-Term Transformer for Online Action Detection","code":"https://github.com/amazon-science/long-short-term-transformer","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":4,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"WOAD","metrics":{"mAP":"67.1"},"uses_additional_data":false,"paper_date":"2020-06-05","paper":"/paper/woad-weakly-supervised-online-action","paper_url":"https://arxiv.org/abs/2006.03732v2","paper_title":"WOAD: Weakly Supervised Online Action Detection in Untrimmed Videos","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"Colar","metrics":{"mAP":"66.9"},"uses_additional_data":false,"paper_date":"2022-03-02","paper":"/paper/colar-effective-and-efficient-online-action","paper_url":"https://arxiv.org/abs/2203.01057v2","paper_title":"Colar: Effective and Efficient Online Action Detection by Consulting Exemplars","code":"https://github.com/vividle/online-action-detection","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":7,"model":"GateHUB (RGB only)","metrics":{"mAP":"66.5"},"uses_additional_data":false,"paper_date":"2022-06-09","paper":"/paper/gatehub-gated-history-unit-with-background-1","paper_url":"https://arxiv.org/abs/2206.04668v1","paper_title":"GateHUB: Gated History Unit with Background Suppression for Online Action Detection","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"OadTR","metrics":{"mAP":"65.2"},"uses_additional_data":false,"paper_date":"2021-06-21","paper":"/paper/oadtr-online-action-detection-with","paper_url":"https://arxiv.org/abs/2106.11149v1","paper_title":"OadTR: Online Action Detection with Transformers","code":"https://github.com/wangxiang1230/OadTR","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":0,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"OadTR-b2","metrics":{"MFLOPs per pred":"1075.7","mAP":"64.5"},"uses_additional_data":false,"paper_date":"2022-01-17","paper":"/paper/continual-transformers-redundancy-free","paper_url":"https://arxiv.org/abs/2201.06268v3","paper_title":"Continual Transformers: Redundancy-Free Attention for Online Inference","code":"https://github.com/lukashedegaard/continual-transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"CoOadTR-b2","metrics":{"MFLOPs per pred":"411.9","mAP":"64.4"},"uses_additional_data":false,"paper_date":"2022-01-17","paper":"/paper/continual-transformers-redundancy-free","paper_url":"https://arxiv.org/abs/2201.06268v3","paper_title":"Continual Transformers: Redundancy-Free Attention for Online Inference","code":"https://github.com/lukashedegaard/continual-transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"OadTR","metrics":{"MFLOPs per pred":"2513.5","mAP":"64.2"},"uses_additional_data":false,"paper_date":"2022-01-17","paper":"/paper/continual-transformers-redundancy-free","paper_url":"https://arxiv.org/abs/2201.06268v3","paper_title":"Continual Transformers: Redundancy-Free Attention for Online Inference","code":"https://github.com/lukashedegaard/continual-transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"OadTR-b1","metrics":{"MFLOPs per pred":"673","mAP":"63.9"},"uses_additional_data":false,"paper_date":"2022-01-17","paper":"/paper/continual-transformers-redundancy-free","paper_url":"https://arxiv.org/abs/2201.06268v3","paper_title":"Continual Transformers: Redundancy-Free Attention for Online Inference","code":"https://github.com/lukashedegaard/continual-transformers","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"TRN","metrics":{"mAP":"62.1"},"uses_additional_data":false,"paper_date":"2018-11-18","paper":"/paper/temporal-recurrent-networks-for-online-action","paper_url":"http://arxiv.org/abs/1811.07391v2","paper_title":"Temporal Recurrent Networks for Online Action Detection","code":"https://github.com/xumingze0308/TRN.pytorch","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"Colar(RGB only)","metrics":{"mAP":"58.6"},"uses_additional_data":false,"paper_date":"2022-03-02","paper":"/paper/colar-effective-and-efficient-online-action","paper_url":"https://arxiv.org/abs/2203.01057v2","paper_title":"Colar: Effective and Efficient Online Action Detection by Consulting Exemplars","code":"https://github.com/vividle/online-action-detection","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":15,"model":"CoOadTR-b1","metrics":{"MFLOPs per pred":"10.6"},"uses_additional_data":false,"paper_date":"2022-01-17","paper":"/paper/continual-transformers-redundancy-free","paper_url":"https://arxiv.org/abs/2201.06268v3","paper_title":"Continual Transformers: Redundancy-Free Attention for Online Inference","code":"https://github.com/lukashedegaard/continual-transformers","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 7,081 of the 9,623 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":9623,"papers_checked":7081,"papers_extracted_not_yet_verified":217,"boards_without_verdict":27,"papers_not_yet_extracted":2325},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+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":5,"rows_with_any_sample_ran":2,"distinct_papers_with_graph_line":4,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":11,"n_unverified":11,"n_samples":22,"n_pointer_only_licence":1,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":11,"n_unverified":12,"n_samples":23,"n_pointer_only_licence":2,"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"}}}