A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation

6D pose estimation of objects is essential for intelligent manufacturing. Current methods mainly place emphasis on the single object’s pose estimation, which limit its use in real-world applications. In this paper, we propose a multi-instance framework of 6D pose estimation for textureless objects i...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Chenrui Wu, Long Chen, Shiqing Wu
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	6D pose estimation metric learning dense correspondences antenna support Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/2a38798e66854dbe80b824a1dcc85322
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:2a38798e66854dbe80b824a1dcc85322
record_format	dspace
spelling	oai:doaj.org-article:2a38798e66854dbe80b824a1dcc853222021-11-25T16:30:59ZA Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation10.3390/app1122105312076-3417https://doaj.org/article/2a38798e66854dbe80b824a1dcc853222021-11-01T00:00:00Zhttps://www.mdpi.com/2076-3417/11/22/10531https://doaj.org/toc/2076-34176D pose estimation of objects is essential for intelligent manufacturing. Current methods mainly place emphasis on the single object’s pose estimation, which limit its use in real-world applications. In this paper, we propose a multi-instance framework of 6D pose estimation for textureless objects in an industrial environment. We use a two-stage pipeline for this purpose. In the detection stage, EfficientDet is used to detect target instances from the image. In the pose estimation stage, the cropped images are first interpolated into a fixed size, then fed into a pseudo-siamese graph matching network to calculate dense point correspondences. A modified circle loss is defined to measure the differences of positive and negative correspondences. Experiments on the antenna support demonstrate the effectiveness and advantages of our proposed method.Chenrui WuLong ChenShiqing WuMDPI AGarticle6D pose estimationmetric learningdense correspondencesantenna supportTechnologyTEngineering (General). Civil engineering (General)TA1-2040Biology (General)QH301-705.5PhysicsQC1-999ChemistryQD1-999ENApplied Sciences, Vol 11, Iss 10531, p 10531 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	6D pose estimation metric learning dense correspondences antenna support Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999
spellingShingle	6D pose estimation metric learning dense correspondences antenna support Technology T Engineering (General). Civil engineering (General) TA1-2040 Biology (General) QH301-705.5 Physics QC1-999 Chemistry QD1-999 Chenrui Wu Long Chen Shiqing Wu A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation
description	6D pose estimation of objects is essential for intelligent manufacturing. Current methods mainly place emphasis on the single object’s pose estimation, which limit its use in real-world applications. In this paper, we propose a multi-instance framework of 6D pose estimation for textureless objects in an industrial environment. We use a two-stage pipeline for this purpose. In the detection stage, EfficientDet is used to detect target instances from the image. In the pose estimation stage, the cropped images are first interpolated into a fixed size, then fed into a pseudo-siamese graph matching network to calculate dense point correspondences. A modified circle loss is defined to measure the differences of positive and negative correspondences. Experiments on the antenna support demonstrate the effectiveness and advantages of our proposed method.
format	article
author	Chenrui Wu Long Chen Shiqing Wu
author_facet	Chenrui Wu Long Chen Shiqing Wu
author_sort	Chenrui Wu
title	A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation
title_short	A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation
title_full	A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation
title_fullStr	A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation
title_full_unstemmed	A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation
title_sort	novel metric-learning-based method for multi-instance textureless objects’ 6d pose estimation
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/2a38798e66854dbe80b824a1dcc85322
work_keys_str_mv	AT chenruiwu anovelmetriclearningbasedmethodformultiinstancetexturelessobjects6dposeestimation AT longchen anovelmetriclearningbasedmethodformultiinstancetexturelessobjects6dposeestimation AT shiqingwu anovelmetriclearningbasedmethodformultiinstancetexturelessobjects6dposeestimation AT chenruiwu novelmetriclearningbasedmethodformultiinstancetexturelessobjects6dposeestimation AT longchen novelmetriclearningbasedmethodformultiinstancetexturelessobjects6dposeestimation AT shiqingwu novelmetriclearningbasedmethodformultiinstancetexturelessobjects6dposeestimation
_version_	1718413125433163776

A Novel Metric-Learning-Based Method for Multi-Instance Textureless Objects’ 6D Pose Estimation

Ejemplares similares