Creates a series of locally position dependent histograms of the values in I.
Inspired by David Lowe's SIFT descriptor. Takes I, divides it into a number of regions,
and creates a histogram for each region. I is divided into approximately equally sized
hyper-rectangular regions so that together these hyper-rectangles cover I. The
hyper-rectangles are actually 'soft', in that each region is actually defined by a
gaussian mask, for details see mask_gaussians. pargmask, parameters to mask_gaussians,
controls details about how the masks are created. Optionally, each value in I may have
associated weight given by weightmask, which should have the same exact dimensions as I.
INPUTS
I - M1xM2x...xMk numeric array
edges - either nbins+1 length vector of quantization bounds, or nbins
pargmask - cell of parameters to mask_gaussians
weightmask - [optional] size(I) numeric array of weights
OUTPUTS
hs - histograms (array of size nbins x nmasks)
EXAMPLE
G = filter_gauss_nD([100 100],[],[],0);
hs = histc_sift( G, 5, {2,.6,.1,0} ); figure(1); im(hs)
DATESTAMP
29-Sep-2005 2:00pm
See also HISTC_1D, MASK_GAUSSIANS, HISTC_SIFT_ND