using neural netwoks and neural trees for vector quantization of images

in this paper we describe the use of neural networks for vector quantization VQ. We first show how a collection of neural units can be used efficienly for VQ encoding , with the neural units performing the bulk of the computation in parallel . We describe an unsupervised learning algorithm for training the neural network vector quantizer . then we extend this algorithm to train neural trees and use them for vector quantization finally we compare these methods to the LBG algorithm in encoding images