{"url":"/sota/action-segmentation-on-50-salads-1","task":{"name":"Action Segmentation","url":"/task/action-segmentation","note":null},"dataset":{"name":"50 Salads","url":"/dataset/50-salads"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Action Segmentation** is a challenging problem in high-level video understanding. In its simplest form, Action Segmentation aims to segment a temporally untrimmed video by time and label each segmented part with one of pre-defined action labels. The results of Action Segmentation can be further used as input to various applications, such as video-to-text and action localization.\n\n\n<span class=\"description-source\">Source: [TricorNet: A Hybrid Temporal Convolutional and Recurrent Network for Video Action Segmentation ](https://arxiv.org/abs/1705.07818)</span>","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":["F1@50%","F1@25%","F1@10%","Acc","Edit"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"F1@50%":"higher","F1@25%":"higher","F1@10%":"higher","Acc":"higher","Edit":null}},"counts":{"rows":28,"rows_with_code":22,"rows_with_paper_page":28,"rows_dated":27,"rows_using_additional_data":3},"rows":[{"rank_in_archive_order":1,"model":"Br-Prompt+ASPnet (RGB, flow, accelerometer)","metrics":{"Acc":"91.4","Edit":"87.5","F1@10%":"92.7","F1@25%":"91.6","F1@50%":"88.5"},"uses_additional_data":true,"paper_date":"2023-01-01","paper":"/paper/aspnet-action-segmentation-with-shared","paper_url":"http://openaccess.thecvf.com//content/CVPR2023/html/van_Amsterdam_ASPnet_Action_Segmentation_With_Shared-Private_Representation_of_Multiple_Data_Sources_CVPR_2023_paper.html","paper_title":"ASPnet: Action Segmentation With Shared-Private Representation of Multiple Data Sources","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"Semantic2Graph","metrics":{"Acc":"88.6","Edit":"89.1","F1@10%":"91.5","F1@25%":"90.2","F1@50%":"87.3"},"uses_additional_data":true,"paper_date":"2022-09-13","paper":"/paper/semantic2graph-graph-based-multi-modal","paper_url":"https://arxiv.org/abs/2209.05653v5","paper_title":"Semantic2Graph: Graph-based Multi-modal Feature Fusion for Action Segmentation in Videos","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"BaFormer","metrics":{"Acc":"89.5","Edit":"84.2","F1@10%":"89.3","F1@25%":"88.4","F1@50%":"83.9"},"uses_additional_data":false,"paper_date":"2024-05-25","paper":"/paper/efficient-temporal-action-segmentation-via","paper_url":"https://arxiv.org/abs/2405.15995v1","paper_title":"Efficient Temporal Action Segmentation via Boundary-aware Query Voting","code":"https://github.com/peiyao-w/baformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"DiffAct","metrics":{"Acc":"88.9","Edit":"85.0","F1@10%":"90.1","F1@25%":"89.2","F1@50%":"83.7"},"uses_additional_data":false,"paper_date":"2023-03-31","paper":"/paper/diffusion-action-segmentation","paper_url":"https://arxiv.org/abs/2303.17959v2","paper_title":"Diffusion Action Segmentation","code":"https://github.com/finspire13/diffact","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":4,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"SF-TMN(ASFormer)","metrics":{"Acc":"89.8","Edit":"84.4","F1@10%":"89.1","F1@25%":"88.0","F1@50%":"82.9"},"uses_additional_data":false,"paper_date":"2023-06-15","paper":"/paper/sf-tmn-slowfast-temporal-modeling-network-for","paper_url":"https://arxiv.org/abs/2306.08859v1","paper_title":"SF-TMN: SlowFast Temporal Modeling Network for Surgical Phase Recognition","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":6,"model":"LTContext","metrics":{"Acc":"87.7","Edit":"83.2","F1@10%":"89.4","F1@25%":"87.7","F1@50%":"82.0"},"uses_additional_data":false,"paper_date":"2023-08-22","paper":"/paper/how-much-temporal-long-term-context-is-needed","paper_url":"https://arxiv.org/abs/2308.11358v2","paper_title":"How Much Temporal Long-Term Context is Needed for Action Segmentation?","code":"https://github.com/ltcontext/ltcontext","n_code_links":1,"syntology":{"n_ran":6,"n_unverified":0,"n_samples":6,"n_pointer_only_licence":6}},{"rank_in_archive_order":7,"model":"UVAST","metrics":{"Acc":"87.4","Edit":"83.9","F1@10%":"89.1","F1@25%":"87.6","F1@50%":"81.7"},"uses_additional_data":false,"paper_date":"2022-09-01","paper":"/paper/unified-fully-and-timestamp-supervised","paper_url":"https://arxiv.org/abs/2209.00638v2","paper_title":"Unified Fully and Timestamp Supervised Temporal Action Segmentation via Sequence to Sequence Translation","code":"https://github.com/boschresearch/uvast","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":8,"model":"Br-Prompt+ASFormer","metrics":{"Acc":"88.1","Edit":"83.8","F1@10%":"89.2","F1@25%":"87.8","F1@50%":"81.3"},"uses_additional_data":true,"paper_date":"2022-03-26","paper":"/paper/bridge-prompt-towards-ordinal-action","paper_url":"https://arxiv.org/abs/2203.14104v1","paper_title":"Bridge-Prompt: Towards Ordinal Action Understanding in Instructional Videos","code":"https://github.com/ttlmh/bridge-prompt","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":9,"model":"EUT","metrics":{"Acc":"87.4","Edit":"82.9","F1@10%":"89.2","F1@25%":"87.5","F1@50%":"81"},"uses_additional_data":false,"paper_date":"2022-05-26","paper":"/paper/efficient-u-transformer-with-boundary-aware","paper_url":"https://arxiv.org/abs/2205.13425v2","paper_title":"Do we really need temporal convolutions in action segmentation?","code":"https://github.com/ddz16/TUT","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"CETNet","metrics":{"Acc":"86.9","Edit":"81.7","F1@10%":"87.6","F1@25%":"86.5","F1@50%":"80.1"},"uses_additional_data":false,"paper_date":"2022-05-19","paper":"/paper/cross-enhancement-transformer-for-action","paper_url":"https://arxiv.org/abs/2205.09445v1","paper_title":"Cross-Enhancement Transformer for Action Segmentation","code":"https://github.com/Wangjhdeveloper/CETNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"DPRN","metrics":{"Acc":"87.2","Edit":"82.0","F1@10%":"87.8","F1@25%":"86.3","F1@50%":"79.4"},"uses_additional_data":false,"paper_date":"2022-05-02","paper":"/paper/maximization-and-restoration-action","paper_url":"https://www.sciencedirect.com/science/article/pii/S003132032200245X","paper_title":"Maximization and restoration: Action segmentation through dilation passing and temporal reconstruction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":12,"model":"ASFormer+ASRF","metrics":{"Acc":"85.9","Edit":"81.9","F1@10%":"85.1","F1@25%":"85.4","F1@50%":"79.3"},"uses_additional_data":false,"paper_date":"2021-10-16","paper":"/paper/asformer-transformer-for-action-segmentation","paper_url":"https://arxiv.org/abs/2110.08568v1","paper_title":"ASFormer: Transformer for Action Segmentation","code":"https://github.com/chinayi/asformer","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"ASRF + HASR","metrics":{"Acc":"83.9","Edit":"81.0","F1@10%":"86.6","F1@25%":"85.7","F1@50%":"78.5"},"uses_additional_data":false,"paper_date":"2021-01-01","paper":"/paper/refining-action-segmentation-with","paper_url":"http://openaccess.thecvf.com//content/ICCV2021/html/Ahn_Refining_Action_Segmentation_With_Hierarchical_Video_Representations_ICCV_2021_paper.html","paper_title":"Refining Action Segmentation With Hierarchical Video Representations","code":"https://github.com/cotton-ahn/hasr_iccv2021","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"ASRF","metrics":{"Acc":"84.5","Edit":"79.3","F1@10%":"84.9","F1@25%":"83.5","F1@50%":"77.3"},"uses_additional_data":false,"paper_date":"2020-07-14","paper":"/paper/alleviating-over-segmentation-errors-by","paper_url":"https://arxiv.org/abs/2007.06866v1","paper_title":"Alleviating Over-segmentation Errors by Detecting Action Boundaries","code":"https://github.com/yiskw713/asrf","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":11,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":15,"model":"ASFormer","metrics":{"Acc":"85.6","Edit":"79.6","F1@10%":"85.1","F1@25%":"83.4","F1@50%":"76.0"},"uses_additional_data":false,"paper_date":"2021-10-16","paper":"/paper/asformer-transformer-for-action-segmentation","paper_url":"https://arxiv.org/abs/2110.08568v1","paper_title":"ASFormer: Transformer for Action Segmentation","code":"https://github.com/chinayi/asformer","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":16,"model":"ETSN","metrics":{"Acc":"82.0","Edit":"78.8","F1@10%":"85.2","F1@25%":"83.9","F1@50%":"75.4"},"uses_additional_data":false,"paper_date":"2021-04-30","paper":"/paper/efficient-two-step-networks-for-temporal","paper_url":"https://www.sciencedirect.com/science/article/pii/S0925231221006998","paper_title":"Efficient Two-Step Networks for Temporal Action Segmentation","code":"https://github.com/lyhisme/ETSN","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"BCN","metrics":{"Acc":"84.4","Edit":"74.3","F1@10%":"82.3","F1@25%":"81.3","F1@50%":"74"},"uses_additional_data":false,"paper_date":null,"paper":"/paper/boundary-aware-cascade-networks-for-temporal","paper_url":"https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/4845_ECCV_2020_paper.php","paper_title":"Boundary-Aware Cascade Networks for Temporal Action Segmentation","code":"https://github.com/MCG-NJU/BCN","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"SSTDA","metrics":{"Acc":"83.2","Edit":"75.8","F1@10%":"83.0","F1@25%":"81.5","F1@50%":"73.8"},"uses_additional_data":false,"paper_date":"2020-03-05","paper":"/paper/action-segmentation-with-joint-self","paper_url":"https://arxiv.org/abs/2003.02824v3","paper_title":"Action Segmentation with Joint Self-Supervised Temporal Domain Adaptation","code":"https://github.com/cmhungsteve/SSTDA","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"C2F-TCN","metrics":{"Acc":"84.9","Edit":"76.4","F1@10%":"84.3","F1@25%":"81.8","F1@50%":"72.6"},"uses_additional_data":false,"paper_date":"2021-05-23","paper":"/paper/coarse-to-fine-multi-resolution-temporal","paper_url":"https://arxiv.org/abs/2105.10859v1","paper_title":"Coarse to Fine Multi-Resolution Temporal Convolutional Network","code":"https://github.com/dipika-singhania/C2F-TCN","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"DA","metrics":{"Acc":"83.2","Edit":"75.2","F1@10%":"82.0","F1@25%":"80.1","F1@50%":"72.5"},"uses_additional_data":false,"paper_date":"2021-04-15","paper":"/paper/action-segmentation-with-mixed-temporal","paper_url":"https://arxiv.org/abs/2104.07461v2","paper_title":"Action Segmentation with Mixed Temporal Domain Adaptation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"MS-TCN++","metrics":{"Acc":"83.7","Edit":"74.3","F1@10%":"80.7","F1@25%":"78.5","F1@50%":"70.1"},"uses_additional_data":false,"paper_date":"2020-06-16","paper":"/paper/ms-tcn-multi-stage-temporal-convolutional-2","paper_url":"https://arxiv.org/abs/2006.09220v2","paper_title":"MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation","code":"https://github.com/sj-li/MS-TCN2","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"G2L (MS-TCN)","metrics":{"Acc":"82.2","Edit":"73.4","F1@10%":"80.3","F1@25%":"78","F1@50%":"69.8"},"uses_additional_data":false,"paper_date":"2021-01-04","paper":"/paper/global2local-efficient-structure-search-for","paper_url":"https://arxiv.org/abs/2101.00910v2","paper_title":"Global2Local: Efficient Structure Search for Video Action Segmentation","code":"https://github.com/ShangHua-Gao/RFNext","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":3}},{"rank_in_archive_order":23,"model":"MS-TCN++(sh)","metrics":{"Acc":" 82.2","Edit":"70.7","F1@10%":"78.7","F1@25%":"76.6","F1@50%":"68.3"},"uses_additional_data":false,"paper_date":"2020-06-16","paper":"/paper/ms-tcn-multi-stage-temporal-convolutional-2","paper_url":"https://arxiv.org/abs/2006.09220v2","paper_title":"MS-TCN++: Multi-Stage Temporal Convolutional Network for Action Segmentation","code":"https://github.com/sj-li/MS-TCN2","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"AUL","metrics":{"Acc":"77.9","Edit":"77.0","F1@10%":"84.4","F1@25%":"81.3","F1@50%":"67.1"},"uses_additional_data":false,"paper_date":"2023-10-22","paper":"/paper/is-weakly-supervised-action-segmentation","paper_url":"https://www.researchgate.net/publication/372548891_Is_Weakly-supervised_Action_Segmentation_Ready_For_Human-Robot_Interaction_No_Let's_Improve_It_With_Action-union_Learning","paper_title":"Is Weakly-supervised Action Segmentation Ready For Human-Robot Interaction? No, Let's Improve It With Action-union Learning","code":"https://github.com/fandulu/DD-Net","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"DTGRM","metrics":{"Acc":"80","Edit":"72","F1@10%":"79.1","F1@25%":"75.9","F1@50%":"66.1"},"uses_additional_data":false,"paper_date":"2020-12-14","paper":"/paper/temporal-relational-modeling-with-self","paper_url":"https://arxiv.org/abs/2012.07508v1","paper_title":"Temporal Relational Modeling with Self-Supervision for Action Segmentation","code":"https://github.com/redwang/DTGRM","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"DS-TCN","metrics":{"Acc":"80.0","Edit":"70.0","F1@10%":"77.0","F1@25%":"74.43","F1@50%":"65.78"},"uses_additional_data":false,"paper_date":"2021-01-19","paper":"/paper/depthwise-separable-temporal-convolutional","paper_url":"https://ieeexplore.ieee.org/document/9320343","paper_title":"Depthwise Separable Temporal Convolutional Network for Action Segmentation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":27,"model":"MS-TCN","metrics":{"Acc":"80.7","Edit":"67.9","F1@10%":"76.3","F1@25%":"74.0","F1@50%":"64.5"},"uses_additional_data":false,"paper_date":"2019-03-05","paper":"/paper/ms-tcn-multi-stage-temporal-convolutional","paper_url":"http://arxiv.org/abs/1903.01945v2","paper_title":"MS-TCN: Multi-Stage Temporal Convolutional Network for Action Segmentation","code":"https://github.com/yabufarha/ms-tcn","n_code_links":2,"syntology":null},{"rank_in_archive_order":28,"model":"TW-FINCH (K=avg/activity)","metrics":{"Acc":"66.5"},"uses_additional_data":false,"paper_date":"2021-03-20","paper":"/paper/temporally-weighted-hierarchical-clustering","paper_url":"https://arxiv.org/abs/2103.11264v4","paper_title":"Temporally-Weighted Hierarchical Clustering for Unsupervised Action Segmentation","code":"https://github.com/ssarfraz/FINCH-CLustering","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":2,"n_samples":7,"n_pointer_only_licence":1}}],"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 6,795 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":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"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":10,"rows_with_any_sample_ran":9,"distinct_papers_with_graph_line":9,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":27,"n_unverified":17,"n_samples":44,"n_pointer_only_licence":14,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":28,"n_unverified":17,"n_samples":45,"n_pointer_only_licence":14,"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"}}}