Similarity assessment between local regions of 3D CAD model based on tree kernel

DOI

10.2991/icimm-15.2015.260How to use a DOI?

Keywords

3D CAD model; local regions; extended feature tree; convolution tree kernel; similarity assessment.

Abstract

To effectively characterize the topology and geometric information of 3D CAD models, trees, graphs and other non-linear data structures have become the important descriptors of 3D CAD models. However, the matching based on tree or graph is still the main bottleneck of 3D CAD model similar evaluation. In this paper, a tree kernels based similar evaluation algorithm is proposed. With extended feature trees as the feature descriptors of local regions, the method uses convolution operation to break down the extended feature tree into sub-structure, and the final evaluation of extended feature trees is achieved by the matching between their sub-structures. The method ensures the polynomial matching time, improves the retrieval accuracy by refining the matching size. Experimental results demonstrate the effectiveness of the algorithm.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Haonan Luo
AU  - Jing Bai
PY  - 2015/07
DA  - 2015/07
TI  - Similarity assessment between local regions of 3D CAD model based on tree kernel
BT  - Proceedings of the 5th International Conference on Information Engineering for Mechanics and Materials
PB  - Atlantis Press
SP  - 1426
EP  - 1429
SN  - 2352-5401
UR  - https://doi.org/10.2991/icimm-15.2015.260
DO  - 10.2991/icimm-15.2015.260
ID  - Luo2015/07
ER  -