Abstract / Description of output
In this paper we deal with the use the tree-structured belief network (TSBN) as a prior model in segmenting a natural image into a number of predefined classes. The TSBN was trained using the EM algorithm based on a set of training label images. The average log likelihood (or bit rate) of a test set of images shows that the learned TSBN is a better model for images than models based on independent blocks of varying sizes. We also analyze the relative advantages obtained by modelling correlations at different length scales in the tree.
Original language | English |
---|---|
Title of host publication | Artificial Neural Networks, 1999. ICANN 99. Ninth International Conference on (Conf. Publ. No. 470) |
Publisher | IET |
Pages | 31-36 |
Number of pages | 5 |
ISBN (Print) | 0-85296-721-7 |
DOIs | |
Publication status | Published - 1 Jan 1999 |
Keywords / Materials (for Non-textual outputs)
- tree-structured belief network
- EM algorithm
- average log likelihood
- image coding
- learning
- image segmentation