LBGS: a smart approach for very large data sets vector quantization

Giuseppe Campobello, Mirko Mantineo, Giuseppe Patanè and Marco Russo

Abstract

In this paper, LBGS, a new parallel/distributed technique for Vector Quantization is presented. It derives from the well known LBG algorithm and has been designed for very complex problems where both large data sets and large codebooks are involved. Several heuristics have been introduced to make it suitable for implementation on parallel/distributed hardware. These lead to a slight deterioration of the quantization error with respect to the serial version but a large improvement in computing efficiency

Keywords: Clustering; Vector quantization; Unsupervised learning; Parallel; Distributed; Learning

[MyLBGS] G. Campobello, M. Mantineo, G. Patanè, and M. Russo. LBGS: a smart approach for very large data sets vector quantization. Signal Processing: Image Communication, vol. 20, Issue1, January 2005, pp. 91-114.

Download: [ps.gz] [pdf]