imresize spuriously causes NaNs with 4d image #763
Description
With a 4-dimensional array:
m = randn(250,202,3,1)
imresize(m, (277, 224, 3, 1))
Running the imresize() multiple times shows NaN's randomly appearing. If I eliminate the fourth dimension, this works just fine:
imresize(m[:,:,:,1], (277, 224, 3))
I assume this is because imresize() is attempting to interpolate across the 3rd axis, and it's reaching out of bounds or something like that. If that is the case, it would be nice if I had a way to specify the dimensions across which I wanted to resize (which I am attempting to do by passing in an explicit tuple of sizes).