[vtkusers] vtkProcrustes Problem

Tim Hutton T.Hutton at eastman.ucl.ac.uk
Mon Dec 22 13:03:51 EST 2003


Jonathan,

Procrustes is for when there is a vertex-by-vertex correspondence between 
the shapes, and is useful for analysing biological objects with similar 
shape. The starting position of the objects doesn't really affect the end 
result - I suspect that the reason your shapes are not aligning correctly 
is because their vertices do not correspond. For example, two surface scans 
of an object will have a very similar shape but their vertices won't 
necessarily match. If this is the case with your data then 
vtkIterativeClosestPointTransform might perform better for you.

Otherwise, I'm not sure what the problem could be. Send me some screenshots 
(off the list)?

Tim.

At 17:50 22/12/2003 +0000, Jonathan.Bailleul wrote:
>Dear all,
>
>I am attempting to use vtkProcrustes filter to align 3d polygonal
>shapes, but it appears that provided alignment is very unsatisfactory in
>my (simple) case. I suspect that initial hand alignment of shapes might
>help a lot, but I do not see how to perform this!
>
>Just kidding: It would be almost easy if I could just move the actors in
>the rendering windows up to satisfying convergence, and then save each
>one onto a vtk file using the current space coordinates of each point.
>
>Would someone have any clue about my problem!
>In advance, thank you very much for your help.
>
>--
>-----------------------------------
>Jonathan BAILLEUL, Doctorant
>GREYC Image - Université de Caen
>http://www.greyc.ismra.fr/~bailleul
>
>_______________________________________________
>This is the private VTK discussion list.
>Please keep messages on-topic. Check the FAQ at: 
><http://public.kitware.com/cgi-bin/vtkfaq>
>Follow this link to subscribe/unsubscribe:
>http://www.vtk.org/mailman/listinfo/vtkusers





More information about the vtkusers mailing list