Loading…
Zero-shot action recognition by clustered representation with redundancy-free features
Zero-shot action recognition (ZSAR) is a practical and challenging issue, which compensates for the shortcomings of existing action recognition by being able to recognize those action classes that don’t have visual representation during training. However, existing zero-shot action recognition doesn’...
Saved in:
Published in: | Machine vision and applications 2023-11, Vol.34 (6), p.116, Article 116 |
---|---|
Main Authors: | , |
Format: | Article |
Language: | English |
Subjects: | |
Citations: | Items that this one cites |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
cited_by | |
---|---|
cites | cdi_FETCH-LOGICAL-c270t-c254f93503ba4768c3a002bafb9b76afc6b6cd3468202ee25465e7f5218d6df13 |
container_end_page | |
container_issue | 6 |
container_start_page | 116 |
container_title | Machine vision and applications |
container_volume | 34 |
creator | Xia, Limin Wen, Xin |
description | Zero-shot action recognition (ZSAR) is a practical and challenging issue, which compensates for the shortcomings of existing action recognition by being able to recognize those action classes that don’t have visual representation during training. However, existing zero-shot action recognition doesn’t focus on the fact that the generated features have many outliers, which harms the recognition. A new method for zero-shot action recognition is proposed, which suppresses this defect by clustered representation with redundancy-free features. In addition, a generative adversarial network (GAN) with gradient penalty is trained to synthesize stable features, solving the problem of data imbalance and alleviating the bottleneck of unstable features generated in existing methods. To reduce the dimension and the subsequent computation, a redundancy-free feature is introduced into the ZSAR. Experiments performed on Olympic Sports, HMDB51, and UCF101 public datasets prove that our method outperforms the state-of-the-art approaches with absolute gains of 1.8%, 0.3%, and 1.7%, respectively, in zero-shot action recognition. |
doi_str_mv | 10.1007/s00138-023-01470-7 |
format | article |
fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_journals_2874638122</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2874638122</sourcerecordid><originalsourceid>FETCH-LOGICAL-c270t-c254f93503ba4768c3a002bafb9b76afc6b6cd3468202ee25465e7f5218d6df13</originalsourceid><addsrcrecordid>eNp9kE1LxDAQhoMouK7-AU8Fz9HJR5P2KItfsOBFPXgJaTrZ7bK2a5Ii---NW8Gbl5lh5nnfgZeQSwbXDEDfRAAmKgpcUGBSA9VHZMak4JRpVR-TGdR5rqDmp-Qsxg0ASK3ljLy9YxhoXA-psC51Q18EdMOq7w5zsy_cdowJA7b5sAsYsU_2cPvq0jrv2rFvbe_21AfEwqNNY6bOyYm324gXv31OXu_vXhaPdPn88LS4XVLHNaRcS-lrUYJorNSqcsIC8Mb6pm60st6pRrlWSFVx4IiZViVqX3JWtar1TMzJ1eS7C8PniDGZzTCGPr80vNJSiYpxnik-US4MMQb0Zhe6Dxv2hoH5yc9M-ZmcnznkZ3QWiUkUM9yvMPxZ_6P6BnrWdDE</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2874638122</pqid></control><display><type>article</type><title>Zero-shot action recognition by clustered representation with redundancy-free features</title><source>Springer Link</source><creator>Xia, Limin ; Wen, Xin</creator><creatorcontrib>Xia, Limin ; Wen, Xin</creatorcontrib><description>Zero-shot action recognition (ZSAR) is a practical and challenging issue, which compensates for the shortcomings of existing action recognition by being able to recognize those action classes that don’t have visual representation during training. However, existing zero-shot action recognition doesn’t focus on the fact that the generated features have many outliers, which harms the recognition. A new method for zero-shot action recognition is proposed, which suppresses this defect by clustered representation with redundancy-free features. In addition, a generative adversarial network (GAN) with gradient penalty is trained to synthesize stable features, solving the problem of data imbalance and alleviating the bottleneck of unstable features generated in existing methods. To reduce the dimension and the subsequent computation, a redundancy-free feature is introduced into the ZSAR. Experiments performed on Olympic Sports, HMDB51, and UCF101 public datasets prove that our method outperforms the state-of-the-art approaches with absolute gains of 1.8%, 0.3%, and 1.7%, respectively, in zero-shot action recognition.</description><identifier>ISSN: 0932-8092</identifier><identifier>EISSN: 1432-1769</identifier><identifier>DOI: 10.1007/s00138-023-01470-7</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer Berlin Heidelberg</publisher><subject>Activity recognition ; Communications Engineering ; Computer Science ; Datasets ; Feature recognition ; Generative adversarial networks ; Image Processing and Computer Vision ; Networks ; Noise ; Original Paper ; Outliers (statistics) ; Pattern Recognition ; Redundancy ; Representations ; Semantics ; Vision systems</subject><ispartof>Machine vision and applications, 2023-11, Vol.34 (6), p.116, Article 116</ispartof><rights>The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c270t-c254f93503ba4768c3a002bafb9b76afc6b6cd3468202ee25465e7f5218d6df13</cites><orcidid>0000-0002-0251-5302</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids></links><search><creatorcontrib>Xia, Limin</creatorcontrib><creatorcontrib>Wen, Xin</creatorcontrib><title>Zero-shot action recognition by clustered representation with redundancy-free features</title><title>Machine vision and applications</title><addtitle>Machine Vision and Applications</addtitle><description>Zero-shot action recognition (ZSAR) is a practical and challenging issue, which compensates for the shortcomings of existing action recognition by being able to recognize those action classes that don’t have visual representation during training. However, existing zero-shot action recognition doesn’t focus on the fact that the generated features have many outliers, which harms the recognition. A new method for zero-shot action recognition is proposed, which suppresses this defect by clustered representation with redundancy-free features. In addition, a generative adversarial network (GAN) with gradient penalty is trained to synthesize stable features, solving the problem of data imbalance and alleviating the bottleneck of unstable features generated in existing methods. To reduce the dimension and the subsequent computation, a redundancy-free feature is introduced into the ZSAR. Experiments performed on Olympic Sports, HMDB51, and UCF101 public datasets prove that our method outperforms the state-of-the-art approaches with absolute gains of 1.8%, 0.3%, and 1.7%, respectively, in zero-shot action recognition.</description><subject>Activity recognition</subject><subject>Communications Engineering</subject><subject>Computer Science</subject><subject>Datasets</subject><subject>Feature recognition</subject><subject>Generative adversarial networks</subject><subject>Image Processing and Computer Vision</subject><subject>Networks</subject><subject>Noise</subject><subject>Original Paper</subject><subject>Outliers (statistics)</subject><subject>Pattern Recognition</subject><subject>Redundancy</subject><subject>Representations</subject><subject>Semantics</subject><subject>Vision systems</subject><issn>0932-8092</issn><issn>1432-1769</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><recordid>eNp9kE1LxDAQhoMouK7-AU8Fz9HJR5P2KItfsOBFPXgJaTrZ7bK2a5Ii---NW8Gbl5lh5nnfgZeQSwbXDEDfRAAmKgpcUGBSA9VHZMak4JRpVR-TGdR5rqDmp-Qsxg0ASK3ljLy9YxhoXA-psC51Q18EdMOq7w5zsy_cdowJA7b5sAsYsU_2cPvq0jrv2rFvbe_21AfEwqNNY6bOyYm324gXv31OXu_vXhaPdPn88LS4XVLHNaRcS-lrUYJorNSqcsIC8Mb6pm60st6pRrlWSFVx4IiZViVqX3JWtar1TMzJ1eS7C8PniDGZzTCGPr80vNJSiYpxnik-US4MMQb0Zhe6Dxv2hoH5yc9M-ZmcnznkZ3QWiUkUM9yvMPxZ_6P6BnrWdDE</recordid><startdate>20231101</startdate><enddate>20231101</enddate><creator>Xia, Limin</creator><creator>Wen, Xin</creator><general>Springer Berlin Heidelberg</general><general>Springer Nature B.V</general><scope>AAYXX</scope><scope>CITATION</scope><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>P5Z</scope><scope>P62</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PTHSS</scope><orcidid>https://orcid.org/0000-0002-0251-5302</orcidid></search><sort><creationdate>20231101</creationdate><title>Zero-shot action recognition by clustered representation with redundancy-free features</title><author>Xia, Limin ; Wen, Xin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c270t-c254f93503ba4768c3a002bafb9b76afc6b6cd3468202ee25465e7f5218d6df13</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Activity recognition</topic><topic>Communications Engineering</topic><topic>Computer Science</topic><topic>Datasets</topic><topic>Feature recognition</topic><topic>Generative adversarial networks</topic><topic>Image Processing and Computer Vision</topic><topic>Networks</topic><topic>Noise</topic><topic>Original Paper</topic><topic>Outliers (statistics)</topic><topic>Pattern Recognition</topic><topic>Redundancy</topic><topic>Representations</topic><topic>Semantics</topic><topic>Vision systems</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Xia, Limin</creatorcontrib><creatorcontrib>Wen, Xin</creatorcontrib><collection>CrossRef</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central</collection><collection>Advanced Technologies & Aerospace Database (1962 - current)</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>ProQuest advanced technologies & aerospace journals</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>Engineering Collection</collection><jtitle>Machine vision and applications</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Xia, Limin</au><au>Wen, Xin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Zero-shot action recognition by clustered representation with redundancy-free features</atitle><jtitle>Machine vision and applications</jtitle><stitle>Machine Vision and Applications</stitle><date>2023-11-01</date><risdate>2023</risdate><volume>34</volume><issue>6</issue><spage>116</spage><pages>116-</pages><artnum>116</artnum><issn>0932-8092</issn><eissn>1432-1769</eissn><abstract>Zero-shot action recognition (ZSAR) is a practical and challenging issue, which compensates for the shortcomings of existing action recognition by being able to recognize those action classes that don’t have visual representation during training. However, existing zero-shot action recognition doesn’t focus on the fact that the generated features have many outliers, which harms the recognition. A new method for zero-shot action recognition is proposed, which suppresses this defect by clustered representation with redundancy-free features. In addition, a generative adversarial network (GAN) with gradient penalty is trained to synthesize stable features, solving the problem of data imbalance and alleviating the bottleneck of unstable features generated in existing methods. To reduce the dimension and the subsequent computation, a redundancy-free feature is introduced into the ZSAR. Experiments performed on Olympic Sports, HMDB51, and UCF101 public datasets prove that our method outperforms the state-of-the-art approaches with absolute gains of 1.8%, 0.3%, and 1.7%, respectively, in zero-shot action recognition.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/s00138-023-01470-7</doi><orcidid>https://orcid.org/0000-0002-0251-5302</orcidid></addata></record> |
fulltext | fulltext |
identifier | ISSN: 0932-8092 |
ispartof | Machine vision and applications, 2023-11, Vol.34 (6), p.116, Article 116 |
issn | 0932-8092 1432-1769 |
language | eng |
recordid | cdi_proquest_journals_2874638122 |
source | Springer Link |
subjects | Activity recognition Communications Engineering Computer Science Datasets Feature recognition Generative adversarial networks Image Processing and Computer Vision Networks Noise Original Paper Outliers (statistics) Pattern Recognition Redundancy Representations Semantics Vision systems |
title | Zero-shot action recognition by clustered representation with redundancy-free features |
url | http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T06%3A13%3A24IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Zero-shot%20action%20recognition%20by%20clustered%20representation%20with%20redundancy-free%20features&rft.jtitle=Machine%20vision%20and%20applications&rft.au=Xia,%20Limin&rft.date=2023-11-01&rft.volume=34&rft.issue=6&rft.spage=116&rft.pages=116-&rft.artnum=116&rft.issn=0932-8092&rft.eissn=1432-1769&rft_id=info:doi/10.1007/s00138-023-01470-7&rft_dat=%3Cproquest_cross%3E2874638122%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c270t-c254f93503ba4768c3a002bafb9b76afc6b6cd3468202ee25465e7f5218d6df13%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2874638122&rft_id=info:pmid/&rfr_iscdi=true |