Vector quantization and clustering: a pyramid approach

Abstract

A multi-resolution K-means clustering method is presented. Starting with a low resolution sample of the input data the K-means algorithm is applied to a sequence of monotonically increasing-resolution samples of the given data. The cluster centers obtained from a low resolution stage are used as initial cluster centers for the next stage which is a higher resolution stage. The idea behind this method is that a good estimation of the initial location of the cluster centers can be obtained through K-means clustering of a sample of the input data. K-means clustering of the entire data with the initial cluster centers estimated by clustering a sample of the input data, reduces the convergence time of the algorithm.