size differences within mesh collections

[Jon Linden]

Hello, I have a set of femur meshes that are not in correspondence and would like to build a shape model.  I wanted to double-check to make sure I am following the appropriate path.

1.  Orient the femurs so that they are in a similar orientation

2.  Choosing one of the meshes at random to serve as a reference, use Parametric non-rigid registration (tutorial 12) and write the projected versions of the meshes into a new set of files.  These files will now be corresponding. 

3.  use the DataCollection class as outlined in tutorial 6 to generate the shape model

Does this sound correct?

Also, how are relative differences in overall size accounted for? For example, a tall person vs a short person may have femurs with exactly the same shape but different size.  We would want them normalized before building a shape model, correct?

Thank you

Jon

[Marcel Luethi]

Hi Jon,

This sounds correct. In case your data is noisy, you may want to work with the fitted mesh instead of the projecteion in step 2, as the projection would also explain the noise.

As discussed previously on this list, the standard gpa method implemented in DataCollection does not standardize the size. Hence you will see the size difference in your models. If you want to normalize those, you can easily do that by changing the gpa method such that it uses a similarity3DLandmarkRegistration instead of the  rigid3DLandmarkRegistration (see https://github.com/unibas-gravis/scalismo/blob/master/src/main/scala/scalismo/statisticalmodel/dataset/DataCollection.scala#L163).

Best regards,

Marcel

--
You received this message because you are subscribed to the Google Groups "scalismo" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scalismo+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/scalismo/a2e5e281-37c5-4870-91fd-1c3a530e8810%40googlegroups.com.

[Jon Linden]

OK great,  thank you Marcel.  

To unsubscribe from this group and stop receiving emails from it, send an email to scal...@googlegroups.com.