<br>This is a classical problem in shape modeling and it is still an active research problem. I am not aware of any vtk implementations.<br>I would recommend the following recently published survey paper:<br><br>"Statistical shape models for 3D medical image segmentation: A review "<br>
<br>Siqi<br><br><div class="gmail_quote">On Thu, Dec 31, 2009 at 5:45 PM, David Doria <span dir="ltr"><<a href="mailto:daviddoria%2Bvtk@gmail.com">daviddoria+vtk@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On Thu, Dec 31, 2009 at 9:57 AM, <<a href="mailto:Stefan.Daenzer@gmail.com">Stefan.Daenzer@gmail.com</a>> wrote:<br>
> Hi,<br>
><br>
> I was looking into vtkPCAAnalysisFilter to do PCA Analysis of a couple of<br>
> training shapes. The training shapes for the vtkPCAAnalysisFilter are<br>
> supposed to have 1-to-1 correspondence among surface points. I was wondering<br>
> if there is a filter in VTK which takes several vtkpolydata objects (without<br>
> 1-to-1 correspondence) as input and generates shapes with corresponding<br>
> points.<br>
><br>
> regards,<br>
><br>
> Stefan<br>
<br>
Short answer - no, not that I know of.<br>
<br>
You can use vtkKdTreePointLocator to find closest points - here is an example:<br>
<a href="http://www.vtk.org/Wiki/VTK/Examples/Filtering/KdTreePointLocator_ClosestPoint" target="_blank">http://www.vtk.org/Wiki/VTK/Examples/Filtering/KdTreePointLocator_ClosestPoint</a><br>
<br>
If you could demonstrate vtkPCAAnalysisFilter here:<br>
<a href="http://www.vtk.org/Wiki/index.php?title=VTK/Examples/vtkPCAAnalysisFilter&action=edit&redlink=1" target="_blank">http://www.vtk.org/Wiki/index.php?title=VTK/Examples/vtkPCAAnalysisFilter&action=edit&redlink=1</a><br>
<br>
that would be much appreciated :)<br>
<br>
It seems to me there would be a lot of heuristics involved with taking<br>
two point sets and finding a 1-to-1 correspondence. When would a point<br>
not have a corresponding point? When there is no point in the other<br>
set close enough? When the closest point in the other set has already<br>
been assigned to correspond to a different point? Do you have any<br>
papers that describe a method to do this? Or did you have an algorithm<br>
in mind?<br>
<br>
Thanks,<br>
<font color="#888888"><br>
David<br>
</font><div><div></div><div class="h5">_______________________________________________<br>
Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
<br>
Visit other Kitware open-source projects at <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
<br>
Please keep messages on-topic and check the VTK FAQ at: <a href="http://www.vtk.org/Wiki/VTK_FAQ" target="_blank">http://www.vtk.org/Wiki/VTK_FAQ</a><br>
<br>
Follow this link to subscribe/unsubscribe:<br>
<a href="http://www.vtk.org/mailman/listinfo/vtkusers" target="_blank">http://www.vtk.org/mailman/listinfo/vtkusers</a><br>
</div></div></blockquote></div><br>