Applying neighborhood consistency for fast clustering and kernel density estimation

Nearest neighborhood consistency is an important concept in statistical pattern recognition, which underlies the well-known k-nearest neighbor method. In this paper, we combine this Idea with kernel density eslimalionhased clustering, and derive thefasl mean shift algorithm (FMS). FMS greatly reduces the complexity of feature space analysis, resulting satisfactotyprecision of classification. More importantly, we show that with FMS algorithm, we are in fact relying on a conceptually novel approach of density estimation, the fast kernel density estimation (FKDE) for clustering. The FKDE combines smooth and non-smooth estimators and (has inherits advantages from both. Asymptotic analysis reveals the approximation of the FKDE to standard kernel density estimator. Data clustering and image segmentation experiments demonstrate the efficiency of FMS.