Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks
Autor(a) principal: | |
---|---|
Data de Publicação: | 2016 |
Outros Autores: | , |
Tipo de documento: | Artigo de conferência |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UNESP |
Texto Completo: | http://dx.doi.org/10.1109/SIBGRAPI.2016.39 http://hdl.handle.net/11449/163000 |
Resumo: | Despite the consistent advances in visual features and other Content-Based Image Retrieval techniques, measuring the similarity among images is still a challenging task for effective image retrieval. In this scenario, similarity learning approaches capable of improving the effectiveness of retrieval in an unsupervised way are indispensable. A novel method, called Cartesian Product of Ranking References (CPRR), is proposed with this objective in this paper. The proposed method uses Cartesian product operations based on rank information for exploiting the underlying structure of datasets. Only subsets of ranked lists are required, demanding low computational efforts. An extensive experimental evaluation was conducted considering various aspects, four public datasets and several image features. Besides effectiveness, experiments were also conducted to assess the efficiency of the method, considering parallel and heterogeneous computing on CPU and GPU devices. The proposed method achieved significant effectiveness gains, including competitive state-of-the-art results on popular benchmarks. |
id |
UNSP_547c7c7b54ddb78ec94a35a23ef405b0 |
---|---|
oai_identifier_str |
oai:repositorio.unesp.br:11449/163000 |
network_acronym_str |
UNSP |
network_name_str |
Repositório Institucional da UNESP |
repository_id_str |
2946 |
spelling |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Taskscontent-based image retrievalunsupervised learningCartesian producteffectivenessefficiencyDespite the consistent advances in visual features and other Content-Based Image Retrieval techniques, measuring the similarity among images is still a challenging task for effective image retrieval. In this scenario, similarity learning approaches capable of improving the effectiveness of retrieval in an unsupervised way are indispensable. A novel method, called Cartesian Product of Ranking References (CPRR), is proposed with this objective in this paper. The proposed method uses Cartesian product operations based on rank information for exploiting the underlying structure of datasets. Only subsets of ranked lists are required, demanding low computational efforts. An extensive experimental evaluation was conducted considering various aspects, four public datasets and several image features. Besides effectiveness, experiments were also conducted to assess the efficiency of the method, considering parallel and heterogeneous computing on CPU and GPU devices. The proposed method achieved significant effectiveness gains, including competitive state-of-the-art results on popular benchmarks.Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)Univ Estadual Paulista UNESP, Dept Stat Appl Math & Comp, Rio Claro, BrazilUniv Estadual Paulista UNESP, Dept Stat Appl Math & Comp, Rio Claro, BrazilFAPESP: 2013/08645-0FAPESP: 2014/04220-8IeeeUniversidade Estadual Paulista (Unesp)Valem, Lucas Pascotti [UNESP]Guimaraes Pedronette, Daniel Carlos [UNESP]IEEE2018-11-26T17:39:43Z2018-11-26T17:39:43Z2016-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject249-256http://dx.doi.org/10.1109/SIBGRAPI.2016.392016 29th Sibgrapi Conference On Graphics, Patterns And Images (sibgrapi). New York: Ieee, p. 249-256, 2016.1530-1834http://hdl.handle.net/11449/16300010.1109/SIBGRAPI.2016.39WOS:000405493800033Web of Sciencereponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPeng2016 29th Sibgrapi Conference On Graphics, Patterns And Images (sibgrapi)info:eu-repo/semantics/openAccess2021-10-23T21:44:31Zoai:repositorio.unesp.br:11449/163000Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T14:28:52.367728Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false |
dc.title.none.fl_str_mv |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
title |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
spellingShingle |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks Valem, Lucas Pascotti [UNESP] content-based image retrieval unsupervised learning Cartesian product effectiveness efficiency |
title_short |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
title_full |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
title_fullStr |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
title_full_unstemmed |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
title_sort |
Unsupervised Similarity Learning through Cartesian Product of Ranking References for Image Retrieval Tasks |
author |
Valem, Lucas Pascotti [UNESP] |
author_facet |
Valem, Lucas Pascotti [UNESP] Guimaraes Pedronette, Daniel Carlos [UNESP] IEEE |
author_role |
author |
author2 |
Guimaraes Pedronette, Daniel Carlos [UNESP] IEEE |
author2_role |
author author |
dc.contributor.none.fl_str_mv |
Universidade Estadual Paulista (Unesp) |
dc.contributor.author.fl_str_mv |
Valem, Lucas Pascotti [UNESP] Guimaraes Pedronette, Daniel Carlos [UNESP] IEEE |
dc.subject.por.fl_str_mv |
content-based image retrieval unsupervised learning Cartesian product effectiveness efficiency |
topic |
content-based image retrieval unsupervised learning Cartesian product effectiveness efficiency |
description |
Despite the consistent advances in visual features and other Content-Based Image Retrieval techniques, measuring the similarity among images is still a challenging task for effective image retrieval. In this scenario, similarity learning approaches capable of improving the effectiveness of retrieval in an unsupervised way are indispensable. A novel method, called Cartesian Product of Ranking References (CPRR), is proposed with this objective in this paper. The proposed method uses Cartesian product operations based on rank information for exploiting the underlying structure of datasets. Only subsets of ranked lists are required, demanding low computational efforts. An extensive experimental evaluation was conducted considering various aspects, four public datasets and several image features. Besides effectiveness, experiments were also conducted to assess the efficiency of the method, considering parallel and heterogeneous computing on CPU and GPU devices. The proposed method achieved significant effectiveness gains, including competitive state-of-the-art results on popular benchmarks. |
publishDate |
2016 |
dc.date.none.fl_str_mv |
2016-01-01 2018-11-26T17:39:43Z 2018-11-26T17:39:43Z |
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/10.1109/SIBGRAPI.2016.39 2016 29th Sibgrapi Conference On Graphics, Patterns And Images (sibgrapi). New York: Ieee, p. 249-256, 2016. 1530-1834 http://hdl.handle.net/11449/163000 10.1109/SIBGRAPI.2016.39 WOS:000405493800033 |
url |
http://dx.doi.org/10.1109/SIBGRAPI.2016.39 http://hdl.handle.net/11449/163000 |
identifier_str_mv |
2016 29th Sibgrapi Conference On Graphics, Patterns And Images (sibgrapi). New York: Ieee, p. 249-256, 2016. 1530-1834 10.1109/SIBGRAPI.2016.39 WOS:000405493800033 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
2016 29th Sibgrapi Conference On Graphics, Patterns And Images (sibgrapi) |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
249-256 |
dc.publisher.none.fl_str_mv |
Ieee |
publisher.none.fl_str_mv |
Ieee |
dc.source.none.fl_str_mv |
Web of Science 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_ |
1808128365753794560 |