NLOOK : a computational attention model for robot vision
Autor(a) principal: | |
---|---|
Data de Publicação: | 2009 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFRGS |
Texto Completo: | http://hdl.handle.net/10183/72579 |
Resumo: | The computational models of visual attention, originally proposed as cognitive models of human attention, nowadays are being used as front-ends to some robotic vision systems, like automatic object recognition and landmark detection. However, these kinds of applications have different requirements from those originally proposed. More specifically, a robotic vision system must be relatively insensitive to 2D similarity transforms of the image, as in-plane translations, rotations, reflections and scales, and it should also select fixation points in scale as well as position. In this paper a new visual attention model, called NLOOK, is proposed. This model is validated through several experiments, which show that it is less sensitive to 2D similarity transforms than other two well known and publicly available visual attention models: NVT and SAFE. Besides, NLOOK can select more accurate fixations than other attention models, and it can select the scales of fixations, too. Thus, the proposed model is a good tool to be used in robot vision systems. |
id |
UFRGS-2_5f1af7b9a6e8b7cd504508c51015aee7 |
---|---|
oai_identifier_str |
oai:www.lume.ufrgs.br:10183/72579 |
network_acronym_str |
UFRGS-2 |
network_name_str |
Repositório Institucional da UFRGS |
repository_id_str |
|
spelling |
Heinen, Milton RobertoEngel, Paulo Martins2013-06-19T01:43:54Z20090104-6500http://hdl.handle.net/10183/72579000733068The computational models of visual attention, originally proposed as cognitive models of human attention, nowadays are being used as front-ends to some robotic vision systems, like automatic object recognition and landmark detection. However, these kinds of applications have different requirements from those originally proposed. More specifically, a robotic vision system must be relatively insensitive to 2D similarity transforms of the image, as in-plane translations, rotations, reflections and scales, and it should also select fixation points in scale as well as position. In this paper a new visual attention model, called NLOOK, is proposed. This model is validated through several experiments, which show that it is less sensitive to 2D similarity transforms than other two well known and publicly available visual attention models: NVT and SAFE. Besides, NLOOK can select more accurate fixations than other attention models, and it can select the scales of fixations, too. Thus, the proposed model is a good tool to be used in robot vision systems.application/pdfengJournal of the Brazilian Computer Society. Porto Alegre. Vol. 15, n. 3 (2009 Sept.), p. 3-17Inteligência artificialVisão computacionalRobot visionVisual attentionSelective attentionFocus of attentionBiomimetic visionNLOOK : a computational attention model for robot visioninfo:eu-repo/semantics/articleinfo:eu-repo/semantics/otherinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UFRGSinstname:Universidade Federal do Rio Grande do Sul (UFRGS)instacron:UFRGSORIGINAL000733068.pdf000733068.pdfTexto completo (inglês)application/pdf3134877http://www.lume.ufrgs.br/bitstream/10183/72579/1/000733068.pdf203d9a1abedd018458cda0c8e8232f68MD51TEXT000733068.pdf.txt000733068.pdf.txtExtracted Texttext/plain54559http://www.lume.ufrgs.br/bitstream/10183/72579/2/000733068.pdf.txtfe5a339d955441953935e848f6e8ba58MD52THUMBNAIL000733068.pdf.jpg000733068.pdf.jpgGenerated Thumbnailimage/jpeg2113http://www.lume.ufrgs.br/bitstream/10183/72579/3/000733068.pdf.jpg301804ed103a45e481c0fc83c14f7e78MD5310183/725792022-02-22 04:50:50.092261oai:www.lume.ufrgs.br:10183/72579Repositório de PublicaçõesPUBhttps://lume.ufrgs.br/oai/requestopendoar:2022-02-22T07:50:50Repositório Institucional da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS)false |
dc.title.pt_BR.fl_str_mv |
NLOOK : a computational attention model for robot vision |
title |
NLOOK : a computational attention model for robot vision |
spellingShingle |
NLOOK : a computational attention model for robot vision Heinen, Milton Roberto Inteligência artificial Visão computacional Robot vision Visual attention Selective attention Focus of attention Biomimetic vision |
title_short |
NLOOK : a computational attention model for robot vision |
title_full |
NLOOK : a computational attention model for robot vision |
title_fullStr |
NLOOK : a computational attention model for robot vision |
title_full_unstemmed |
NLOOK : a computational attention model for robot vision |
title_sort |
NLOOK : a computational attention model for robot vision |
author |
Heinen, Milton Roberto |
author_facet |
Heinen, Milton Roberto Engel, Paulo Martins |
author_role |
author |
author2 |
Engel, Paulo Martins |
author2_role |
author |
dc.contributor.author.fl_str_mv |
Heinen, Milton Roberto Engel, Paulo Martins |
dc.subject.por.fl_str_mv |
Inteligência artificial Visão computacional |
topic |
Inteligência artificial Visão computacional Robot vision Visual attention Selective attention Focus of attention Biomimetic vision |
dc.subject.eng.fl_str_mv |
Robot vision Visual attention Selective attention Focus of attention Biomimetic vision |
description |
The computational models of visual attention, originally proposed as cognitive models of human attention, nowadays are being used as front-ends to some robotic vision systems, like automatic object recognition and landmark detection. However, these kinds of applications have different requirements from those originally proposed. More specifically, a robotic vision system must be relatively insensitive to 2D similarity transforms of the image, as in-plane translations, rotations, reflections and scales, and it should also select fixation points in scale as well as position. In this paper a new visual attention model, called NLOOK, is proposed. This model is validated through several experiments, which show that it is less sensitive to 2D similarity transforms than other two well known and publicly available visual attention models: NVT and SAFE. Besides, NLOOK can select more accurate fixations than other attention models, and it can select the scales of fixations, too. Thus, the proposed model is a good tool to be used in robot vision systems. |
publishDate |
2009 |
dc.date.issued.fl_str_mv |
2009 |
dc.date.accessioned.fl_str_mv |
2013-06-19T01:43:54Z |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/other |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10183/72579 |
dc.identifier.issn.pt_BR.fl_str_mv |
0104-6500 |
dc.identifier.nrb.pt_BR.fl_str_mv |
000733068 |
identifier_str_mv |
0104-6500 000733068 |
url |
http://hdl.handle.net/10183/72579 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.ispartof.pt_BR.fl_str_mv |
Journal of the Brazilian Computer Society. Porto Alegre. Vol. 15, n. 3 (2009 Sept.), p. 3-17 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFRGS instname:Universidade Federal do Rio Grande do Sul (UFRGS) instacron:UFRGS |
instname_str |
Universidade Federal do Rio Grande do Sul (UFRGS) |
instacron_str |
UFRGS |
institution |
UFRGS |
reponame_str |
Repositório Institucional da UFRGS |
collection |
Repositório Institucional da UFRGS |
bitstream.url.fl_str_mv |
http://www.lume.ufrgs.br/bitstream/10183/72579/1/000733068.pdf http://www.lume.ufrgs.br/bitstream/10183/72579/2/000733068.pdf.txt http://www.lume.ufrgs.br/bitstream/10183/72579/3/000733068.pdf.jpg |
bitstream.checksum.fl_str_mv |
203d9a1abedd018458cda0c8e8232f68 fe5a339d955441953935e848f6e8ba58 301804ed103a45e481c0fc83c14f7e78 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFRGS - Universidade Federal do Rio Grande do Sul (UFRGS) |
repository.mail.fl_str_mv |
|
_version_ |
1801224793479970816 |