Title :
MDL-based design of vector quantizers
Author :
Bischof, Horst ; Leonardis, Ales
Author_Institution :
Pattern Recognition & Image Process. Group, Wien Univ., Austria
Abstract :
We develop a framework for vector quantization networks based on the minimum description length (MDL) principle. This MDL framework is used to derive conditions for the removal of superfluous units from the network. We design a computationally efficient algorithm for finding the optimal number of reference vectors as well as their positions. We illustrate our approach on 2D clustering problems and present applications on image coding
Keywords :
image coding; minimisation; neural nets; probability; unsupervised learning; vector quantisation; 2D clustering; image coding; minimisation; minimum description length; neural nets; probability distribution; superfluous unit removal; unsupervised learning; vector quantization; Clustering algorithms; Distortion measurement; Encoding; Image coding; Neural networks; Pattern recognition; Read only memory; Stochastic processes; Unsupervised learning; Vector quantization;
Conference_Titel :
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-8186-8512-3
DOI :
10.1109/ICPR.1998.711293
