File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Improving action localization by progressive cross-stream cooperation

TitleImproving action localization by progressive cross-stream cooperation
Authors
KeywordsCategorization
Recognition: Detection
Retrieval
Video Analytics
Issue Date2019
Citation
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2019, v. 2019-June, p. 12008-12017 How to Cite?
AbstractSpatio-temporal action localization consists of three levels of tasks: Spatial localization, action classification, and temporal segmentation. In this work, we propose a new Progressive Cross-stream Cooperation (PCSC) framework to iterative improve action localization results and generate better bounding boxes for one stream (i.e., Flow/RGB) by leveraging both region proposals and features from another stream (i.e., RGB/Flow) in an iterative fashion. Specifically, we first generate a larger set of region proposals by combining the latest region proposals from both streams, from which we can readily obtain a larger set of labelled training samples to help learn better action detection models. Second, we also propose a new message passing approach to pass information from one stream to another stream in order to learn better representations, which also leads to better action detection models. As a result, our iterative framework progressively improves action localization results at the frame level. To improve action localization results at the video level, we additionally propose a new strategy to train class-specific actionness detectors for better temporal segmentation, which can be readily learnt by using the training samples around temporal boundaries. Comprehensive experiments on two benchmark datasets UCF-101-24 and J-HMDB demonstrate the effectiveness of our newly proposed approaches for spatio-temporal action localization in realistic scenarios.
Persistent Identifierhttp://hdl.handle.net/10722/321877
ISSN
2023 SCImago Journal Rankings: 10.331
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorSu, Rui-
dc.contributor.authorOuyang, Wanli-
dc.contributor.authorZhou, Luping-
dc.contributor.authorXu, Dong-
dc.date.accessioned2022-11-03T02:22:04Z-
dc.date.available2022-11-03T02:22:04Z-
dc.date.issued2019-
dc.identifier.citationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2019, v. 2019-June, p. 12008-12017-
dc.identifier.issn1063-6919-
dc.identifier.urihttp://hdl.handle.net/10722/321877-
dc.description.abstractSpatio-temporal action localization consists of three levels of tasks: Spatial localization, action classification, and temporal segmentation. In this work, we propose a new Progressive Cross-stream Cooperation (PCSC) framework to iterative improve action localization results and generate better bounding boxes for one stream (i.e., Flow/RGB) by leveraging both region proposals and features from another stream (i.e., RGB/Flow) in an iterative fashion. Specifically, we first generate a larger set of region proposals by combining the latest region proposals from both streams, from which we can readily obtain a larger set of labelled training samples to help learn better action detection models. Second, we also propose a new message passing approach to pass information from one stream to another stream in order to learn better representations, which also leads to better action detection models. As a result, our iterative framework progressively improves action localization results at the frame level. To improve action localization results at the video level, we additionally propose a new strategy to train class-specific actionness detectors for better temporal segmentation, which can be readily learnt by using the training samples around temporal boundaries. Comprehensive experiments on two benchmark datasets UCF-101-24 and J-HMDB demonstrate the effectiveness of our newly proposed approaches for spatio-temporal action localization in realistic scenarios.-
dc.languageeng-
dc.relation.ispartofProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition-
dc.subjectCategorization-
dc.subjectRecognition: Detection-
dc.subjectRetrieval-
dc.subjectVideo Analytics-
dc.titleImproving action localization by progressive cross-stream cooperation-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/CVPR.2019.01229-
dc.identifier.scopuseid_2-s2.0-85078810201-
dc.identifier.volume2019-June-
dc.identifier.spage12008-
dc.identifier.epage12017-
dc.identifier.isiWOS:000542649305064-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats