I was wondering if it is possible to concatenate two dense deformation fields in ITK python or SimpleITK.
B = T2 ( T1 ( A(x) ) )
In which T1, and T2 are both dense deformation fields (both with size ofImage A and with vector type voxels), is it possible to derive
B = T ( A(x) )
which would be concatenation of T1 and T2.
I know there is a CompositeTransform class but does AddTransform, add the deformation fields or it applies the first one and then the second one.
My reason for concatenation of dense fields is that in my registration framework I predict a dense deformation field at each iteration, but I don’t want to resample the images to avoid interpolation artifacts as a result of multiple resampling.