SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously

Detalhes bibliográficos
Autor(a) principal: Aragao, Dunfrey
Data de Publicação: 2020
Outros Autores: Nascimento, Tiago, Mondini, Adriano [UNESP], Paiva, A. C., Conci, A., Braz, G., Almeida, JDS, Fernandes, LAF
Tipo de documento: Artigo de conferência
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://hdl.handle.net/11449/209189
Resumo: One of the fundamental dilemmas of mobile robotics is the use of sensory information to locate an agent in geographic space. In this paper, we developed a global relocation system to predict the robot's position and avoid unforeseen actions from a monocular image, which we named SpaceYNet. We incorporated Inception layers to symmetric layers of down-sampling and upsampling to solve depth-scene and 6-DoF estimation simultaneously. Also, we compared SpaceYNet to PoseNet - a state of the art in robot pose regression using CNN - in order to evaluate it. The comparison comprised one public dataset and one created in a broad indoor environment. SpaceYNet showed higher accuracy in global percentages when compared to PoseNet.
id UNSP_10c7a576b30c9917d7a39e301d9e3538
oai_identifier_str oai:repositorio.unesp.br:11449/209189
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression SimultaneouslyDatasetdepth-sceneposeregressionrobotOne of the fundamental dilemmas of mobile robotics is the use of sensory information to locate an agent in geographic space. In this paper, we developed a global relocation system to predict the robot's position and avoid unforeseen actions from a monocular image, which we named SpaceYNet. We incorporated Inception layers to symmetric layers of down-sampling and upsampling to solve depth-scene and 6-DoF estimation simultaneously. Also, we compared SpaceYNet to PoseNet - a state of the art in robot pose regression using CNN - in order to evaluate it. The comparison comprised one public dataset and one created in a broad indoor environment. SpaceYNet showed higher accuracy in global percentages when compared to PoseNet.Univ Fed Paraiba, Joao Pessoa, Paraiba, BrazilUniv Estadual Paulista, Sao Paulo, BrazilUniv Estadual Paulista, Sao Paulo, BrazilIeeeUniv Fed ParaibaUniversidade Estadual Paulista (Unesp)Aragao, DunfreyNascimento, TiagoMondini, Adriano [UNESP]Paiva, A. C.Conci, A.Braz, G.Almeida, JDSFernandes, LAF2021-06-25T11:50:57Z2021-06-25T11:50:57Z2020-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject217-222Proceedings Of The 2020 International Conference On Systems, Signals And Image Processing (iwssip), 27th Edition. New York: Ieee, p. 217-222, 2020.2157-8672http://hdl.handle.net/11449/209189WOS:000615731300038Web of Sciencereponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPengProceedings Of The 2020 International Conference On Systems, Signals And Image Processing (iwssip), 27th Editioninfo:eu-repo/semantics/openAccess2021-10-23T19:23:38Zoai:repositorio.unesp.br:11449/209189Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462021-10-23T19:23:38Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
title SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
spellingShingle SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
Aragao, Dunfrey
Dataset
depth-scene
pose
regression
robot
title_short SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
title_full SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
title_fullStr SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
title_full_unstemmed SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
title_sort SpaceYNet: A Novel Approach to Pose and Depth-Scene Regression Simultaneously
author Aragao, Dunfrey
author_facet Aragao, Dunfrey
Nascimento, Tiago
Mondini, Adriano [UNESP]
Paiva, A. C.
Conci, A.
Braz, G.
Almeida, JDS
Fernandes, LAF
author_role author
author2 Nascimento, Tiago
Mondini, Adriano [UNESP]
Paiva, A. C.
Conci, A.
Braz, G.
Almeida, JDS
Fernandes, LAF
author2_role author
author
author
author
author
author
author
dc.contributor.none.fl_str_mv Univ Fed Paraiba
Universidade Estadual Paulista (Unesp)
dc.contributor.author.fl_str_mv Aragao, Dunfrey
Nascimento, Tiago
Mondini, Adriano [UNESP]
Paiva, A. C.
Conci, A.
Braz, G.
Almeida, JDS
Fernandes, LAF
dc.subject.por.fl_str_mv Dataset
depth-scene
pose
regression
robot
topic Dataset
depth-scene
pose
regression
robot
description One of the fundamental dilemmas of mobile robotics is the use of sensory information to locate an agent in geographic space. In this paper, we developed a global relocation system to predict the robot's position and avoid unforeseen actions from a monocular image, which we named SpaceYNet. We incorporated Inception layers to symmetric layers of down-sampling and upsampling to solve depth-scene and 6-DoF estimation simultaneously. Also, we compared SpaceYNet to PoseNet - a state of the art in robot pose regression using CNN - in order to evaluate it. The comparison comprised one public dataset and one created in a broad indoor environment. SpaceYNet showed higher accuracy in global percentages when compared to PoseNet.
publishDate 2020
dc.date.none.fl_str_mv 2020-01-01
2021-06-25T11:50:57Z
2021-06-25T11:50:57Z
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 Proceedings Of The 2020 International Conference On Systems, Signals And Image Processing (iwssip), 27th Edition. New York: Ieee, p. 217-222, 2020.
2157-8672
http://hdl.handle.net/11449/209189
WOS:000615731300038
identifier_str_mv Proceedings Of The 2020 International Conference On Systems, Signals And Image Processing (iwssip), 27th Edition. New York: Ieee, p. 217-222, 2020.
2157-8672
WOS:000615731300038
url http://hdl.handle.net/11449/209189
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv Proceedings Of The 2020 International Conference On Systems, Signals And Image Processing (iwssip), 27th Edition
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 217-222
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_ 1803650335875530752

Registros relacionados