Grasp detection via visual rotation object detection and point cloud spatial feature scoring

Accurately detecting the appropriate grasp configurations is the central task for the robot to grasp an object. Existing grasp detection methods usually overlook the depth image or only regard it as a two-dimensional distance image, which makes it difficult to capture the three-dimensional structura...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jie Wang, Shuxiao Li
Formato:	article
Lenguaje:	EN
Publicado:	SAGE Publishing 2021
Materias:	Electronics TK7800-8360 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/c4b6fa3a6d994d4c844f213fa4596b03
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:c4b6fa3a6d994d4c844f213fa4596b03
record_format	dspace
spelling	oai:doaj.org-article:c4b6fa3a6d994d4c844f213fa4596b032021-11-16T01:03:23ZGrasp detection via visual rotation object detection and point cloud spatial feature scoring1729-881410.1177/17298814211055577https://doaj.org/article/c4b6fa3a6d994d4c844f213fa4596b032021-11-01T00:00:00Zhttps://doi.org/10.1177/17298814211055577https://doaj.org/toc/1729-8814Accurately detecting the appropriate grasp configurations is the central task for the robot to grasp an object. Existing grasp detection methods usually overlook the depth image or only regard it as a two-dimensional distance image, which makes it difficult to capture the three-dimensional structural characteristics of target object. In this article, we transform the depth image to point cloud and propose a two-stage grasp detection method based on candidate grasp detection from RGB image and spatial feature rescoring from point cloud. Specifically, we first adopt the recently proposed high-performance rotation object detection method for aerial images, named R3Det, to grasp detection task, obtaining the candidate grasp boxes and their appearance scores. Then, point clouds within each candidate grasp box are normalized and evaluated to get the point cloud quality scores, which are fused with the established point cloud quantity scoring model to obtain spatial scores. Finally, appearance scores and their corresponding spatial scores are combined to output high-quality grasp detection results. The proposed method effectively fuses three types of grasp scoring modules, thus is called Score Fusion Grasp Net. Besides, we propose and adopt top-k grasp metric to effectively reflect the success rate of algorithm in actual grasp execution. Score Fusion Grasp Net obtains 98.5% image-wise accuracy and 98.1% object-wise accuracy on Cornell Grasp Dataset, which exceeds the performances of state-of-the-art methods. We also use the robotic arm to conduct physical grasp experiments on 15 kinds of household objects and 11 kinds of adversarial objects. The results show that the proposed method still has a high success rate when facing new objects.Jie WangShuxiao LiSAGE PublishingarticleElectronicsTK7800-8360Electronic computers. Computer scienceQA75.5-76.95ENInternational Journal of Advanced Robotic Systems, Vol 18 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Electronics TK7800-8360 Electronic computers. Computer science QA75.5-76.95
spellingShingle	Electronics TK7800-8360 Electronic computers. Computer science QA75.5-76.95 Jie Wang Shuxiao Li Grasp detection via visual rotation object detection and point cloud spatial feature scoring
description	Accurately detecting the appropriate grasp configurations is the central task for the robot to grasp an object. Existing grasp detection methods usually overlook the depth image or only regard it as a two-dimensional distance image, which makes it difficult to capture the three-dimensional structural characteristics of target object. In this article, we transform the depth image to point cloud and propose a two-stage grasp detection method based on candidate grasp detection from RGB image and spatial feature rescoring from point cloud. Specifically, we first adopt the recently proposed high-performance rotation object detection method for aerial images, named R3Det, to grasp detection task, obtaining the candidate grasp boxes and their appearance scores. Then, point clouds within each candidate grasp box are normalized and evaluated to get the point cloud quality scores, which are fused with the established point cloud quantity scoring model to obtain spatial scores. Finally, appearance scores and their corresponding spatial scores are combined to output high-quality grasp detection results. The proposed method effectively fuses three types of grasp scoring modules, thus is called Score Fusion Grasp Net. Besides, we propose and adopt top-k grasp metric to effectively reflect the success rate of algorithm in actual grasp execution. Score Fusion Grasp Net obtains 98.5% image-wise accuracy and 98.1% object-wise accuracy on Cornell Grasp Dataset, which exceeds the performances of state-of-the-art methods. We also use the robotic arm to conduct physical grasp experiments on 15 kinds of household objects and 11 kinds of adversarial objects. The results show that the proposed method still has a high success rate when facing new objects.
format	article
author	Jie Wang Shuxiao Li
author_facet	Jie Wang Shuxiao Li
author_sort	Jie Wang
title	Grasp detection via visual rotation object detection and point cloud spatial feature scoring
title_short	Grasp detection via visual rotation object detection and point cloud spatial feature scoring
title_full	Grasp detection via visual rotation object detection and point cloud spatial feature scoring
title_fullStr	Grasp detection via visual rotation object detection and point cloud spatial feature scoring
title_full_unstemmed	Grasp detection via visual rotation object detection and point cloud spatial feature scoring
title_sort	grasp detection via visual rotation object detection and point cloud spatial feature scoring
publisher	SAGE Publishing
publishDate	2021
url	https://doaj.org/article/c4b6fa3a6d994d4c844f213fa4596b03
work_keys_str_mv	AT jiewang graspdetectionviavisualrotationobjectdetectionandpointcloudspatialfeaturescoring AT shuxiaoli graspdetectionviavisualrotationobjectdetectionandpointcloudspatialfeaturescoring
_version_	1718426811332820992

Grasp detection via visual rotation object detection and point cloud spatial feature scoring

Ejemplares similares