Novel VQ with constraints on the quantization error distribution

In this paper, we motivate and introduce a novel vector quantization (VQ) scheme for distributing the quantization error among the quantized features of a continuous feature vector in a predefined manner. This is done by defining ratios between the individual quantization errors of the features and shaping the Voronoi cells accordingly. In a series of experiments we show that the novel approach is capable of either distributing the quantization error equally among the dimensions or realize an arbitrary distribution.

• Type: Journal
• Year: 2009
• Where: ICMCS
• Authors: Joachim Schenk, Frank Wallhoff, Gerhard Rigoll
• Keywords: Continuous Feature Vector, ICMCS 2009, Individual Quantization Errors, Multimedia, Quantization Error