2D Sketch Based Recognition of 3D freeform Shape by Using the RBF Neural Network

Qin, S. F.; Sun, Guangmin; Wright, D. K.; Lim, S.; Khan, U.; Mao, C.

2D Sketch Based Recognition of 3D freeform Shape by Using the RBF Neural Network

Files

119-126.pdf (264.33 KB)

Date

2005

Authors

Qin, S. F.
Sun, Guangmin
Wright, D. K.
Lim, S.
Khan, U.
Mao, C.

Publisher

The Eurographics Association

Abstract

This paper presents a novel free-form surface recognition method from 2D freehand sketching. The approach is based on the Radial basis function (RBF), an artificial intelligence technique. A simple three-layered network has been designed and constructed. After training and testing with two types of surfaces (four sided boundary surfaces and four close section surfaces), it has been shown that the method is useful in freeform surface recognition. The testing results are very satisfactory.

        @inproceedings{:10.2312/SBM/SBM05/119-126
,
booktitle = {Eurographics Workshop on Sketch-Based Interfaces and Modeling
},
editor = {Joaquim Armando Pires Jorge and Takeo Igarashi
},
title = {{2D Sketch Based Recognition of 3D freeform Shape by Using the RBF Neural Network
}},
author = {Qin, S. F. and 
Sun, Guangmin and 
Wright, D. K. and 
Lim, S. and 
Khan, U. and 
Mao, C.
},
year = {2005
},
publisher = {The Eurographics Association
},
ISSN = {1812-3503
},
ISBN = {3-905673-30-4
},
DOI = {/10.2312/SBM/SBM05/119-126
}
}