NF: add version of Paul Ivanov's slice viewer

[matthew-brett]

@ivanov - many thanks for this - it's very nice, and very useful.

The thing I really miss though, is having cross-hairs showing me the
corresponding current point on all three panels.

I think there's an example here:

http://marsbar.sourceforge.net/_images/roi_view.png

I experimented for a while trying to put the crosshairs on, but it was beyond
my blit fu. Any pointers as to how to do that? I mean, put up the
cross-hairs, different on every movement of the slices, but still be able to
blit for speed?

[yarikoptic]

hm, I thought that such high level functionality as visualization was previously thought worth keeping outside?

[matthew-brett]

Yes, indeed, but then I found I was using it a lot and thought it
would be fairly easy to drop in.

[matthew-brett]

Anyone with matplotlib expertise have a look at this one?

[larsoner]

I might take a look at this to make working on parrec easier, but if
someone else has more mpl experience and/or free time feel free to go for
it instead.

[larsoner]

Done:

matthew-brett#5

While we're at it, it might be worth adding a .plot() method to the image files so that people can easily inspect their data. I don't know the code base well enough to know what layer of abstraction that should live in, though.

[larsoner]

Figured I'd work on this before parrec, since it should facilitate the parrec development :)

[matthew-brett]

Merged your changes, rebasing on master now.

[larsoner]

Great, let me know when you get back to this post-release and I'm happy to review changes you make or go through some iterations with you.

[larsoner]

Should close this given nilearn has something similar, or you think it's still worth pursuing here?

http://nilearn.github.io/building_blocks/plotting.html

[matthew-brett]

Sorry to be lazy - but at a first pass these all looked static - so you can't click around the image?

[larsoner]

It was dynamic last time I used it, clicking around the image should work as should linking two images together, etc. IIRC everything was working correctly when we last looked at it, we just shelved it due to the release and concerns that it should be part of the resampling PR, e.g.:

matthew-brett#5 (comment)

[matthew-brett]

Sorry - I mean the nilearn stuff is static - no? If that is true, this PR still does something we need and that we don't have.

[larsoner]

That's true. Do you think it's worth trying to port our changes over to their code instead of partially duplicating effort? That might be a pain, though, I guess.

How's the resampling coming along? :)

[matthew-brett]

Hi Eric. @choldgraf just caught me in the corridor asking about this one.

I would like to get this merged real-soon-now (TM). I need this stuff too. I think we can sort out the resampling axis stuff later. Do you have some time to rebase this? I've got quite a bit of time in the next few days to do final review.

[larsoner]

That was quick, nicely done @choldgraf :) I should be able to look tomorrow.

[larsoner]

Closing for #404