Generate random vectors in PCA subspace. Used to generate random vectors from the subspace spanned by the first k principal components. The points generated come from the gaussian distribution from within the subspace. Can optionally generate points on the subspace that are also on a hypershpere centered on the origin. This may be useful if the original data points were all from a hypershpere -- for example they were normalized via imNormalize. Set the optional hypershpere flag to 1 to generate points only on the hypersphere. USAGE Xr = pcaRandVec( U, mu, vars, k, n, [hypershpere], [show] ) INPUTS U - returned by pca.m mu - returned by pca.m vars - returned by pca.m k - number of principal coordinates to use n - number of points to generate hypershpere - [0] generate points on hypersphere (see above) show - [1] figure to use for display (no display if == 0) OUTPUTS Xr - resulting randomly generated vectors EXAMPLE See also PCA IMNORMALIZE Piotr's Computer Vision Matlab Toolbox Version 2.0 Copyright 2014 Piotr Dollar. [pdollar-at-gmail.com] Licensed under the Simplified BSD License [see external/bsd.txt]