On noisy source vector quantization via a subspace constrained mean shift algorithm

Abstract

The subspace constrained mean shift (SCMS) algorithm is an iterative method for finding an underlying manifold associated with an intrinsically low dimensional data set embedded in a high dimensional space. We investigate the application of the SCMS algorithm to the problem of noisy source vector quantization where the clean source needs to be estimated from its noisy observation before quantizing with an optimal vector quantizer. We demonstrate that an SCMS-based preprocessing step can be effective for sources that have intrinsically low dimensionality in situations where clean source samples are unavailable and the system design relies only on noisy source samples for training.