Cargando…

A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes

Object detection and grasp detection are essential for unmanned systems working in cluttered real-world environments. Detecting grasp configurations for each object in the scene would enable reasoning manipulations. However, finding the relationships between objects and grasp configurations is still...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Yang, Xie, Lihua, Li, Yuheng, Li, Yuan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9986287/
https://www.ncbi.nlm.nih.gov/pubmed/36890968
http://dx.doi.org/10.3389/fncom.2023.1110889
_version_ 1784901131342708736
author Zhang, Yang
Xie, Lihua
Li, Yuheng
Li, Yuan
author_facet Zhang, Yang
Xie, Lihua
Li, Yuheng
Li, Yuan
author_sort Zhang, Yang
collection PubMed
description Object detection and grasp detection are essential for unmanned systems working in cluttered real-world environments. Detecting grasp configurations for each object in the scene would enable reasoning manipulations. However, finding the relationships between objects and grasp configurations is still a challenging problem. To achieve this, we propose a novel neural learning approach, namely SOGD, to predict a best grasp configuration for each detected objects from an RGB-D image. The cluttered background is first filtered out via a 3D-plane-based approach. Then two separate branches are designed to detect objects and grasp candidates, respectively. The relationship between object proposals and grasp candidates are learned by an additional alignment module. A series of experiments are conducted on two public datasets (Cornell Grasp Dataset and Jacquard Dataset) and the results demonstrate the superior performance of our SOGD against SOTA methods in predicting reasonable grasp configurations “from a cluttered scene.”
format Online
Article
Text
id pubmed-9986287
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-99862872023-03-07 A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes Zhang, Yang Xie, Lihua Li, Yuheng Li, Yuan Front Comput Neurosci Neuroscience Object detection and grasp detection are essential for unmanned systems working in cluttered real-world environments. Detecting grasp configurations for each object in the scene would enable reasoning manipulations. However, finding the relationships between objects and grasp configurations is still a challenging problem. To achieve this, we propose a novel neural learning approach, namely SOGD, to predict a best grasp configuration for each detected objects from an RGB-D image. The cluttered background is first filtered out via a 3D-plane-based approach. Then two separate branches are designed to detect objects and grasp candidates, respectively. The relationship between object proposals and grasp candidates are learned by an additional alignment module. A series of experiments are conducted on two public datasets (Cornell Grasp Dataset and Jacquard Dataset) and the results demonstrate the superior performance of our SOGD against SOTA methods in predicting reasonable grasp configurations “from a cluttered scene.” Frontiers Media S.A. 2023-02-20 /pmc/articles/PMC9986287/ /pubmed/36890968 http://dx.doi.org/10.3389/fncom.2023.1110889 Text en Copyright © 2023 Zhang, Xie, Li and Li. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Zhang, Yang
Xie, Lihua
Li, Yuheng
Li, Yuan
A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
title A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
title_full A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
title_fullStr A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
title_full_unstemmed A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
title_short A neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
title_sort neural learning approach for simultaneous object detection and grasp detection in cluttered scenes
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9986287/
https://www.ncbi.nlm.nih.gov/pubmed/36890968
http://dx.doi.org/10.3389/fncom.2023.1110889
work_keys_str_mv AT zhangyang aneurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT xielihua aneurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT liyuheng aneurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT liyuan aneurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT zhangyang neurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT xielihua neurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT liyuheng neurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes
AT liyuan neurallearningapproachforsimultaneousobjectdetectionandgraspdetectioninclutteredscenes