{"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/leveraging-triplet-loss-for-unsupervised","title":"Leveraging triplet loss for unsupervised action segmentation","arxiv_id":"2304.06403","date":"2023-04-13","proceeding":null,"authors":["E. Bueno-Benito","B. Tura","M. Dimiccoli"],"abstract":"In this paper, we propose a novel fully unsupervised framework that learns action representations suitable for the action segmentation task from the single input video itself, without requiring any training data. Our method is a deep metric learning approach rooted in a shallow network with a triplet loss operating on similarity distributions and a novel triplet selection strategy that effectively models temporal and semantic priors to discover actions in the new representational space. Under these circumstances, we successfully recover temporal boundaries in the learned action representations with higher quality compared with existing unsupervised approaches. The proposed method is evaluated on two widely used benchmark datasets for the action segmentation task and it achieves competitive performance by applying a generic clustering algorithm on the learned representations.","url_abs":"https://arxiv.org/abs/2304.06403v2","url_pdf":"https://arxiv.org/pdf/2304.06403v2.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":"leveraging-triplet-loss-for-unsupervised","repo_url":"https://github.com/elenabbbuenob/tsa-actionseg","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"action-segmentation","task_name":"Action Segmentation"},{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":null,"task_name":"Triplet"},{"task_slug":"unsupervised-action-segmentation","task_name":"Unsupervised Action Segmentation"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[{"method_slug":"triplet-loss","method_name":"Triplet Loss"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-segmentation-on-breakfast-1","task":"Action Segmentation","dataset":"Breakfast","model":"TSA (FINCH)","rank_in_archive_order":34,"of":37,"metrics":{"Acc":"65.1","mIoU":"52.1"},"uses_additional_data":false},{"leaderboard":"/sota/action-segmentation-on-breakfast-1","task":"Action Segmentation","dataset":"Breakfast","model":"TSA (Kmeans)","rank_in_archive_order":35,"of":37,"metrics":{"Acc":"63.7","F1":"58","mIoU":"53.3"},"uses_additional_data":false},{"leaderboard":"/sota/action-segmentation-on-breakfast-1","task":"Action Segmentation","dataset":"Breakfast","model":"TSA (Spectral)","rank_in_archive_order":36,"of":37,"metrics":{"Acc":"63.2","F1":"57.8","mIoU":"52.7"},"uses_additional_data":false},{"leaderboard":"/sota/action-segmentation-on-youtube-inria","task":"Action Segmentation","dataset":"Youtube INRIA Instructional","model":"TSA (FINCH)","rank_in_archive_order":1,"of":2,"metrics":{"Acc":"62.4","F1":"54.7"},"uses_additional_data":false},{"leaderboard":"/sota/action-segmentation-on-youtube-inria","task":"Action Segmentation","dataset":"Youtube INRIA Instructional","model":"TSA (Kmeans)","rank_in_archive_order":2,"of":2,"metrics":{"Acc":"59.7","F1":"55.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2304.06403","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}