[Insight-users] Non-uniform sampling in image registration

[Eve]

Hi users,

Would the following extensions be sensible for non-uniform sampling in the
image domain for intensity-based image registration?

1. allow for input of a mesh in the metric involved (i.e. octree
partitioning of the image) 
2. iterate over each element in the mesh and draw N samples in each
partition (such that small partitions will have more samples drawn than
those in large partitions)

If so, what is the best/most efficient representation of the mesh/partition
in this scenario? And are there any classes already available in ITK can be
used for this task?


Thanks,
Eve

-- 
View this message in context: http://www.nabble.com/Non-uniform-sampling-in-image-registration-tp15370356p15370356.html
Sent from the ITK - Users mailing list archive at Nabble.com.