An open-source implementation of meanshift clustering implementation for MATLAB/Octave.
This is an improved version of the meanshift implementation appears in MATLAB File Exchange. The support for arbitary kernel is added.
Copyright 2015 Han Gong gong@fedoraproject.org, University of East Anglia
Copyright 2006 Bryan Feldman
References:
Fukunaga, Keinosuke, and Larry D. Hostetler. "The estimation of the gradient of a density function, with applications in pattern recognition." Information Theory, IEEE Transactions on 21.1 (1975): 32-40.
MATLAB/Octave Code: see demo.m for demonstration.
Results:
