Carter Francis

@pc494 you can take a stab at it just to get the tests passing, otherwise I can do it as well. There are some larger problems with the `phase_cross_correlation` function...

This should wait until #868 is merged as there are some functions which are deprecated here.

One thing we discussed at the developers meeting the other day was including additional N-D signals which were the same size as the navigation signal. I will share my thoughts...

@pc494 @rbjorge I should hopefully fix this bug while creating a BaseVector Signal in hyperspy and then restructuring the DiffractionVectors class over the next couple of months. The idea with...

> The main issue one has to keep in mind is that copying the data from memory or disk to the GPU and back is expensive, so ideally one does...

@pc494 I think that this one should be ready to go. I'm going to hold off on adding any GPU enabled stuff for now.

@dnjohnstone I was planning on doing this quickly today. I've got most of the changes done, just needed to find the time to clean it up some more.

After looking at this a little bit more I think that this might better be added to skimage than into pyxem. I will start a PR there and see if...

This PR should be closed with the addition into scikit-image below. All further correlation methods should just call the cross_correlate_masked function from skimage. https://github.com/scikit-image/scikit-image/pull/5573 Holding off for now until this...

Just a note that with the revival of https://github.com/hyperspy/hyperspy/pull/2972 we should maybe reconsider how to integrate this. I think that the idea of a separate plugin library for hyperspy is...