Closed ralfHielscher closed 1 year ago
What is the current behavior?
I use the following code in my principal geodesic analysis function... to find a best-fit orientation dispersion tensor... and the resultant tensor (or it's eigenvectors) can be used to define a fibre.
oriRef = mean(ori);
% compute the projection of orientations onto the three dimensional % tangential space centered about the mean t = log(ori, oriRef,'left');
% covariance matrix as tensor T = tensor(t*t,'rank',2);
Hi Zach,
i think this was working before with fibre.fit
. The problems start when your orientations are not locally concentrated. Then, due to symmetry, my original and also your approach will not give the right result.
If have updated the function in b2f754b696c33c13d070cc48f3daada850769706 to work also for globally spread orientations.
I would be happy if you could test it.
Ralf.
fibre.fit