Abstract
The task of 3D shape classification is to assign a set of unordered shapes into pre-tagged classes with class labels. In this paper, we present a 3D shape classifier approach based on supervision of the learning of point spatial distributions. We first extract the low-level features by characterising the point spatial density distributions, and train one feed-forward neural network to learn these features by examples. The Konstanz shape database was chosen as the test database to evaluate the accuracy rate of classification. We also compared this classifier to the k nearest neighbours classifier for 3D shapes.
Original language | English |
---|---|
Pages (from-to) | 134-143 |
Number of pages | 10 |
Journal | International Journal of Computer Applications in Technology |
Volume | 38 |
Issue number | 1-3 |
DOIs | |
State | Published - 2010 |
Externally published | Yes |
Keywords
- 3D shape classification
- 3D shape classifier
- Neural network supervision
- Point spatial distributions