You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As noted in SyneRBI/SIRF#680, using the symmetric version of f3d causes a rotation of the registered image and the resulting transformations (deformations and displacements).
This can be seen by comparing outputs of:
reg_f3d -flo test.nii -ref test.nii -sym
and
reg_f3d -flo test.nii -ref test.nii
When testing with an image where nx!=ny!=nz, the symmetric results look even stranger. I wonder if this implies that some images are being iterated across in the wrong order. When square, this results in a rotation, and when non-square, this results in a striding error.
I'm attaching two screenshots. The first screenshot, with nx==ny!=nz, shows a rotation. The second, with nx!=ny!=nz shows the distortion.
For both screenshots, test is the input, and outSym and outNoSym are the registered images, with and without symmetric, respectively.
Square
Non-square
The text was updated successfully, but these errors were encountered:
As noted in SyneRBI/SIRF#680, using the symmetric version of f3d causes a rotation of the registered image and the resulting transformations (deformations and displacements).
This can be seen by comparing outputs of:
and
When testing with an image where nx!=ny!=nz, the symmetric results look even stranger. I wonder if this implies that some images are being iterated across in the wrong order. When square, this results in a rotation, and when non-square, this results in a striding error.
I'm attaching two screenshots. The first screenshot, with nx==ny!=nz, shows a rotation. The second, with nx!=ny!=nz shows the distortion.
For both screenshots, test is the input, and outSym and outNoSym are the registered images, with and without symmetric, respectively.
Square
Non-square
The text was updated successfully, but these errors were encountered: