File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Slow Motion Matters: A Slow Motion Enhanced Network for Weakly Supervised Temporal Action Localization

TitleSlow Motion Matters: A Slow Motion Enhanced Network for Weakly Supervised Temporal Action Localization
Authors
KeywordsFeature extraction
Location awareness
Motion segmentation
slow motion
Sports
Task analysis
temporal action localization
Training
Videos
Weakly-supervised learning
Issue Date2022
Citation
IEEE Transactions on Circuits and Systems for Video Technology, 2022 How to Cite?
AbstractWeakly supervised temporal action localization (WTAL) aims to localize actions in untrimmed videos with only weak supervision information (e.g., video-level labels). Most existing models handle all input videos with a fixed temporal scale. However, such models are not sensitive to actions whose pace of the movements is different from the “normal” speed, especially slow-motion action instances, which complete the movements with a much slower speed than their counterparts with a “normal” speed. Here arises the slow-motion blurred issue: It is hard to explore salient slow-motion information from videos at normal speed. In this paper, we propose a novel framework termed Slow Motion Enhanced Network (SMEN) to improve the ability of a WTAL network by compensating its sensitivity on slow-motion action segments. The proposed SMEN comprises a Mining module and a Localization module. The mining module generates mask to mine slow-motion-related features by utilizing the relationships between the normal motion and slow motion; while the localization module leverages the mined slow-motion features as complementary information to improve the temporal action localization results. Our proposed framework can be easily adapted by existing WTAL networks and enable them be more sensitive to slow-motion actions. Extensive experiments on three benchmarks are conducted, which demonstrate the high performance of our proposed framework.
Persistent Identifierhttp://hdl.handle.net/10722/322011
ISSN
2023 Impact Factor: 8.3
2023 SCImago Journal Rankings: 2.299
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorSun, Weiqi-
dc.contributor.authorSu, Rui-
dc.contributor.authorYu, Qian-
dc.contributor.authorXu, Dong-
dc.date.accessioned2022-11-03T02:23:00Z-
dc.date.available2022-11-03T02:23:00Z-
dc.date.issued2022-
dc.identifier.citationIEEE Transactions on Circuits and Systems for Video Technology, 2022-
dc.identifier.issn1051-8215-
dc.identifier.urihttp://hdl.handle.net/10722/322011-
dc.description.abstractWeakly supervised temporal action localization (WTAL) aims to localize actions in untrimmed videos with only weak supervision information (<italic>e.g</italic>., video-level labels). Most existing models handle all input videos with a fixed temporal scale. However, such models are not sensitive to actions whose pace of the movements is different from the &#x201C;normal&#x201D; speed, especially slow-motion action instances, which complete the movements with a much slower speed than their counterparts with a &#x201C;normal&#x201D; speed. Here arises the slow-motion blurred issue: It is hard to explore salient slow-motion information from videos at normal speed. In this paper, we propose a novel framework termed Slow Motion Enhanced Network (SMEN) to improve the ability of a WTAL network by compensating its sensitivity on slow-motion action segments. The proposed SMEN comprises a Mining module and a Localization module. The mining module generates mask to mine slow-motion-related features by utilizing the relationships between the normal motion and slow motion; while the localization module leverages the mined slow-motion features as complementary information to improve the temporal action localization results. Our proposed framework can be easily adapted by existing WTAL networks and enable them be more sensitive to slow-motion actions. Extensive experiments on three benchmarks are conducted, which demonstrate the high performance of our proposed framework.-
dc.languageeng-
dc.relation.ispartofIEEE Transactions on Circuits and Systems for Video Technology-
dc.subjectFeature extraction-
dc.subjectLocation awareness-
dc.subjectMotion segmentation-
dc.subjectslow motion-
dc.subjectSports-
dc.subjectTask analysis-
dc.subjecttemporal action localization-
dc.subjectTraining-
dc.subjectVideos-
dc.subjectWeakly-supervised learning-
dc.titleSlow Motion Matters: A Slow Motion Enhanced Network for Weakly Supervised Temporal Action Localization-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TCSVT.2022.3201540-
dc.identifier.scopuseid_2-s2.0-85137608298-
dc.identifier.eissn1558-2205-
dc.identifier.isiWOS:000911746000026-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats