3D CAD Model Retrieval Based on Oriented Bounding Box and Normals Distribution

Article Preview

Abstract:

Shape Distribution (SD) describes the global feature of 3D CAD model which can be used to distinguish one model from others. But sometimes, different models may have the similar SD features. In this case SD is not suitable for 3D CAD model retrieval. Although two models have the same SD features, their oriented bounding box (OBB) and normals distribution may differ from each other. In this paper, a new algorithm based on OBB and normals distribution is proposed to tackle this problem. SD feature is used for initial retrieval, then OBB and normals distribution are used for fine retrieval. Experimental results indicate the efficiency and feasibility of the proposed method.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

2324-2327

Citation:

Online since:

January 2014

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] Chen Q, Yu Y.M. 3D CAD Model Retrieval Based on Feature Fusion. Advanced Materials Research, Vol 765, (2013), pp.316-319.

DOI: 10.4028/www.scientific.net/amr.765-767.316

Google Scholar

[2] Osada R, Funkhouser T, Chazelle B. Shape distributions . ACM Transactions on Graphics, Vol 21, no 4, (2002), pp.807-832.

DOI: 10.1145/571647.571648

Google Scholar

[3] Wu Y.H., Tian L., Li C.G. High Efficient Methods of Content based 3D Model Retrieval. Chinese Journal of Mechanical Engineering, Vol 26, no 1, (2013), pp.248-256.

DOI: 10.3901/cjme.2013.02.248

Google Scholar

[4] ESB LIB on https: /engineering. purdue. edu/cdesign/wp/?page_id=1386.

Google Scholar

[5] Shilane P, Min P, Kazhdan M, et al. The princeton shape benchmark. Shape Modeling Applications, (2004), pp.167-178.

DOI: 10.1109/smi.2004.1314504

Google Scholar