Unsupervised distance learning by reciprocal kNN distance for image retrieval
Autor(a) principal: | |
---|---|
Data de Publicação: | 2014 |
Outros Autores: | , , |
Tipo de documento: | Artigo de conferência |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1145/2578726.2578770 http://hdl.handle.net/11449/171552 |
Resumo: | This paper presents a novel unsupervised learning approach that takes into account the intrinsic dataset structure, which is represented in terms of the reciprocal neighborhood references found in different ranked lists. The proposed Reciprocal kNN Distance defines a more effective distance between two images, and is used to improve the effectiveness of image retrieval systems. Several experiments were conducted for different image retrieval tasks involving shape, color, and texture descriptors. The proposed approach is also evaluated on multimodal retrieval tasks, considering visual and textual descriptors. Experimental results demonstrate the effectiveness of proposed approach. The Reciprocal kNN Distance yields better results in terms of effectiveness than various state-of-the-art algorithms. Copyright © 2014 ACM. |
id |
UNSP_b3d2498abceada40a3d0ec68ac076138 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/171552 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Unsupervised distance learning by reciprocal kNN distance for image retrievalContent-based image retrievalUnsupervised distance learningThis paper presents a novel unsupervised learning approach that takes into account the intrinsic dataset structure, which is represented in terms of the reciprocal neighborhood references found in different ranked lists. The proposed Reciprocal kNN Distance defines a more effective distance between two images, and is used to improve the effectiveness of image retrieval systems. Several experiments were conducted for different image retrieval tasks involving shape, color, and texture descriptors. The proposed approach is also evaluated on multimodal retrieval tasks, considering visual and textual descriptors. Experimental results demonstrate the effectiveness of proposed approach. The Reciprocal kNN Distance yields better results in terms of effectiveness than various state-of-the-art algorithms. Copyright © 2014 ACM.Advanced Micro DevicesDepartment of Statistic, Applied Math. and Computing, Universidade Estadual Paulista (UNESP), Rio-Claro, SP, 13506-900RECOD Lab., Institute of Computing, University of Campinas (UNICAMP), Campinas, SP, 13083-852Advanced Technologies, SAMSUNG Research Institute, Campinas, SP, 13097-104Department of Exact Sciences, University of Feira de Santana (UEFS), Feira de Santana, BA, 44036-900Department of Statistic, Applied Math. and Computing, Universidade Estadual Paulista (UNESP), Rio-Claro, SP, 13506-900Universidade Estadual Paulista (Unesp)Universidade Estadual de Campinas (UNICAMP)Advanced Technologies, SAMSUNG Research InstitutePedronette, Daniel C. G. [UNESP]Penatti, Otávio A. B.Calumby, Rodrigo T.Da S. Torres, Ricardo2018-12-11T16:55:48Z2018-12-11T16:55:48Z2014-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject345-352http://dx.doi.org/10.1145/2578726.2578770ICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014, p. 345-352.http://hdl.handle.net/11449/17155210.1145/2578726.25787702-s2.0-84899769548Scopusreponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014info:eu-repo/semantics/openAccess2021-10-23T21:46:58Zoai:repositorio.unesp.br:11449/171552Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T19:55:52.108638Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
title |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
spellingShingle |
Unsupervised distance learning by reciprocal kNN distance for image retrieval Pedronette, Daniel C. G. [UNESP] Content-based image retrieval Unsupervised distance learning |
title_short |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
title_full |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
title_fullStr |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
title_full_unstemmed |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
title_sort |
Unsupervised distance learning by reciprocal kNN distance for image retrieval |
author |
Pedronette, Daniel C. G. [UNESP] |
author_facet |
Pedronette, Daniel C. G. [UNESP] Penatti, Otávio A. B. Calumby, Rodrigo T. Da S. Torres, Ricardo |
author_role |
author |
author2 |
Penatti, Otávio A. B. Calumby, Rodrigo T. Da S. Torres, Ricardo |
author2_role |
author author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) Universidade Estadual de Campinas (UNICAMP) Advanced Technologies, SAMSUNG Research Institute |
dc.contributor.author.fl_str_mv |
Pedronette, Daniel C. G. [UNESP] Penatti, Otávio A. B. Calumby, Rodrigo T. Da S. Torres, Ricardo |
dc.subject.por.fl_str_mv |
Content-based image retrieval Unsupervised distance learning |
topic |
Content-based image retrieval Unsupervised distance learning |
description |
This paper presents a novel unsupervised learning approach that takes into account the intrinsic dataset structure, which is represented in terms of the reciprocal neighborhood references found in different ranked lists. The proposed Reciprocal kNN Distance defines a more effective distance between two images, and is used to improve the effectiveness of image retrieval systems. Several experiments were conducted for different image retrieval tasks involving shape, color, and texture descriptors. The proposed approach is also evaluated on multimodal retrieval tasks, considering visual and textual descriptors. Experimental results demonstrate the effectiveness of proposed approach. The Reciprocal kNN Distance yields better results in terms of effectiveness than various state-of-the-art algorithms. Copyright © 2014 ACM. |
publishDate |
2014 |
dc.date.none.fl_str_mv |
2014-01-01 2018-12-11T16:55:48Z 2018-12-11T16:55:48Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/conferenceObject |
format |
conferenceObject |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://dx.doi.org/10.1145/2578726.2578770 ICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014, p. 345-352. http://hdl.handle.net/11449/171552 10.1145/2578726.2578770 2-s2.0-84899769548 |
url |
http://dx.doi.org/10.1145/2578726.2578770 http://hdl.handle.net/11449/171552 |
identifier_str_mv |
ICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014, p. 345-352. 10.1145/2578726.2578770 2-s2.0-84899769548 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
ICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
345-352 |
dc.source.none.fl_str_mv |
Scopus reponame:Repositório Institucional da UNESP instname:Universidade Estadual Paulista (UNESP) instacron:UNESP |
instname_str |
Universidade Estadual Paulista (UNESP) |
instacron_str |
UNESP |
institution |
UNESP |
reponame_str |
Repositório Institucional da UNESP |
collection |
Repositório Institucional da UNESP |
repository.name.fl_str_mv |
Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP) |
repository.mail.fl_str_mv |
|
_version_ |
1808129140545552384 |