Classified vector quantization using direction classification

Abstract

In this paper, an effective classifier based on the direction of vectors for classified vector quantization is proposed. The new classifier uses dimensionality-independent distortion measure based on inner product to determine the similarity between vectors. The distortion measure is simple and enough to classify various edge types other than single and straight line types, which limit the size of image block to a rather small size. Simulation results show that the technique can achieve good perceptual quality and edge intergrity at low bit rate with reduced coding complexity.