Algorithms for fast vector quantization

Abstract

This paper shows that if one is willing to relax the requirement of finding the true nearest neighbor, it is possible to achieve significant improvements in running time and at only a very small loss in the performance of the vector quantizer. The authors present three algorithms for nearest neighbor searching: standard and priority k-d tree search algorithms and a neighborhood graph search algorithm in which a directed graph is constructed for the point set and edges join neighboring points.<>