Fast weighted median. Computes the weighted median of a set of samples. http://en.wikipedia.org/wiki/Weighted_median "A weighted median of a sample is the 50% weighted percentile." For matrices computes median along each column (or dimension dim). If all weights are equal to 1 gives identical results to median. USAGE m = medianw( x, w, [dim] ) INPUTS x - vector or array of samples w - vector or array of weights dim - dimension along which to compute median OUTPUTS m - weighted median value of x EXAMPLE - simple toy example x=[1 2 3]; w=[1 1 5]; medianw(x,w) EXAMPLE - comparison to median n=randi(100); m=randi(100); x=rand(n,m); w=ones(n,m); m1=median(x); m2=medianw(x,w); assert(isequal(m1,m2)) See also median Piotr's Computer Vision Matlab Toolbox Version 3.24 Copyright 2014 Piotr Dollar. [pdollar-at-gmail.com] Licensed under the Simplified BSD License [see external/bsd.txt]