function [featureVector] = makeColumnFeatureTexture(img, type, numpartitions)
% Version 0.1, Feb 2006.
% This code is copyrighted by Ashutosh Saxena, Stanford University
% It is available for non-commercial use only. For commericial use, please
% contact the authors.
% Any publication or report resulting from Use of this code, should cite:
% Learning Depth from Single Monocular Images, Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng, NIPS 2005.
% 3-D Depth Reconstruction from a Single Still Image, Ashutosh Saxena, Sung H. Chung, Andrew Y. Ng, To appear in IJCV 2007.
% For more info and dataset, visit: http://ai.stanford.edu/~asaxena/learningdepth/
% For a better version of the algorithm, also visit: 
% http://ai.stanford.edu/~asaxena/reconstruction3d/



persistent L3 E3 S3 L5 L7 gMask		% the Laws' masks

global columnWidth nGridCol nGridRow rowWidth
global scale2Factor
global nLaw nDim nHarris

if nargin < 3
	numpartitions = 40;
end
if nargin < 2
	type = 'texture';
end

nGCol = nGridCol;
nGRow = nGridRow;

% Assume that the image is correctly oriented.
imheight = size(img, 1);
imwidth  = size(img, 2);
% Step sizes in x and y
stepwidth  = imwidth/nGridCol;
stepheight = imheight/nGridRow;


H2 = calculateFilterBanks(img, type);

%columnfeatures = zeros(numpartitions, 2*nLaw+3*nHarris);
columnfeatures = zeros(1, numpartitions, nLaw);
% column step size
colstepsize = round(imwidth / numpartitions);
for step=1:numpartitions
	colLeft  = round( (step - 1)*colstepsize + 1 );
	colRight = round( step*colstepsize );
	if colRight > imwidth
		colRight = imwidth;
	end
	if nLaw
		% This old version of the code uses 1 patch for column feature vector, modify to match it with one in NIPS and IJCV paper.
		columnfeatures(1, step, 1:nLaw) = sum( sum( H2( :, colLeft:colRight, 1:nLaw), 2), 1);
%		columnfeatures(step, nLaw+(1:nLaw)) = sum( sum( H4( :, colLeft:colRight, 1:nLaw), 2), 1);
	end
	if nHarris
		% Do Harris here if need to.
	end
end

%featureVectorTmp = zeros( nGRow, nGCol, 2*nLaw+3*nHarris);
featureVectorTmp = zeros( nGRow, nGCol, nLaw);

for g=1:nGRow
	for c=1:nGCol
		choice = floor((c*stepwidth)/colstepsize) + 1;
		if choice > numpartitions
			choice = numpartitions;
		end
		featureVectorTmp(g,c,:) = columnfeatures(1, choice,:);
	end
end
 
featureVector = featureVectorTmp;


return;