[Insight-users] Writing mesh to Metafile

[Dominique Zosso]

Hi Julien,

Thank you for your quick response.

As mentioned, the itkIndexedTriangleMesh is being developed and maintained
by Tobias Heiman, so I think it's up to him to contribute whenever he
considers this appropriate.

Your second point is a more straightforward solution for my application. In
fact, the CellLinks (so the only information contained in the "Improved"
mesh) can easily be rebuilt from the normal Boundary Assignments contained
in the classical mesh. Therefore, there is no critical data loss when
casting the itkIndexedTriangleMesh into a normal itkMesh for saving, as the
reverse casting can be done when reading the mesh lateron.

I'll try this at once.

Thank you,
Dominique

----- Original Message ----- 
From: "Julien Jomier" <jjomier at cs.unc.edu>
To: "Dominique Zosso" <zosso at tele.ucl.ac.be>
Cc: "Insight Users" <insight-users at itk.org>
Sent: Thursday, December 08, 2005 6:18 PM
Subject: Re: [Insight-users] Writing mesh to Metafile


> Hi Dominique,
>
> We are having the same issue with the SimplexMesh class which derives
> from itk::Mesh.
> This is a problem with classes deriving from itk::Mesh and because of
> some cast issues there is no way to write them out at this point using
> the MeshSpatialObject.
>
> So you have two options:
>
> 1) If you plan to make the itkIndexedTriangleMesh class part of ITK, I'd
>   be happy to create a new itkIndexedTriangleMeshSpatialObject so you
> can write this type of object out.
>
> 2) If you can convert the itkIndexedTriangleMesh to an itkMesh structure
> then you can use the current writer.
>
> I'll investigate if there is a better way to make the writer more
> generic in the future.
>
> let us know if that would work for you,
>
> Julien
>