Model-based work-piece localization with salient feature selection
Zhu, Wenjun; Qin, Zhengke; Wang, Peng; Qiao, Hong
2013-12
Conference NameRobotics and Biomimetics (ROBIO), 2013 IEEE International Conference on
Conference Date12-14 Dec. 2013
Conference PlaceShenzhen, China
Abstract
This paper presents a model-based work-piece localization method with salient feature selection. Model-based localization is suitable for work-piece which is one kind of the typical 3D rigid objects with less texture. However, localization based on 3D model will cause high failure rate in heavily cluttered scenes. We propose a new model-based localization method, which is integrated with salient feature selection. Two different models: 3D model and training images are used, and the salient feature selection procedure extracts the regions which may contain the objects potentially. Experiments demonstrate the effectiveness of the proposed method.
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/14767
Collection复杂系统管理与控制国家重点实验室_机器人理论与应用
AffiliationInstitute of Automaton, Chinese Academy of Science
Recommended Citation
GB/T 7714
Zhu, Wenjun,Qin, Zhengke,Wang, Peng,et al. Model-based work-piece localization with salient feature selection[C],2013.
Files in This Item: Download All
File Name/Size DocType Version Access License
workpiece-localizati(999KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhu, Wenjun]'s Articles
[Qin, Zhengke]'s Articles
[Wang, Peng]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhu, Wenjun]'s Articles
[Qin, Zhengke]'s Articles
[Wang, Peng]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhu, Wenjun]'s Articles
[Qin, Zhengke]'s Articles
[Wang, Peng]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: workpiece-localization.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.