[Insight-developers] Re: [Insight-users] DeformableModelSimplexMesh

Leila Baghdadi baghdadi at sickkids.ca
Thu Aug 24 10:43:53 EDT 2006


Hi Alex.

As Master Luis mentioned, Thomas started this originally and I followed
up to get this working for my own research and of course the research
community. 

I have done a fair bit of studies about simplex meshes and deformable
models, mostly the work of Herve Delingette of INRIA France.

I would be more than glad to assist in any way that I can including
coding of various pieces should you require any help.

Please do not hesitate to e-mail me (even better the developers list)


Leila

On Thu, 2006-24-08 at 08:44 -0400, Luis Ibanez wrote:
> Hi Alex,
> 
> What you are undertaking is a very interesting effort.
> 
> We will be happy to assist you in the coding of these new Mesh
> classes.  Please note that ITK has a backward compatibility
> policy that states that we will keep in the toolkit classes
> that have been made available in any previous release.
> 
> To be consistent with these policy, when we add new classes
> that may be replacement for existing classes, we still keep
> the original ones in the toolkit. That means that instead
> of modifying the code of the deformable filters, what we
> would like to do is to add new versions that are appropriate
> for your new Mesh structure. In this way, both the old filter
> and the new filters will be available to users.
> 
> You probably noticed that this approach was already used for
> the Simplex mesh. The original deformable model filter was
> written for the itkMesh, and when Thomas Boettger kindly
> contributed the Simplex mesh to the toolkit, he also provided
> a customized version of the Deformable Model algorithm, that
> was adapted to work with the Simplex mesh.
> 
> 
> 
> Do you have a design for these classes ?
> 
> We will be very interested in looking at the hierarchy
> that you are planning to implement.
> 
> 
> One resource that you could use is to add a page in the ITK
> Wiki (anybody has write access to it) describing the details
> of your design and implementation.  We do this for any major
> modification to the toolkit.
> 
> The page were you can add this proposal is:
> 
>   http://www.itk.org/Wiki/ITK_Oversight_Committee#Proposals
> 
> 
> Also, since you are planning to contribute code to the
> toolkit, we should probably move this discussion to the
> developers list. You can subscribe to the list at the
> following link:
> 
> http://www.itk.org/mailman/listinfo/insight-developers
> 
> 
> You are also welcome to joing the weekly phone conferences
> that are held on Fridays at 1:00pm EST. You will find
> information about this at:
> 
> http://www.itk.org/Wiki/Telephone_Conference_Agendas_and_Minutes
> 
> 
> Please let us know if you have any questions,
> 
> 
>    Thanks
> 
> 
>      Luis
> 
> 
> 
> -----------------------
> alex. gouaillard wrote:
> > Dear leila, dear all.
> > 
> > FYI, we are almost ready (one month) to release a new mesh structure in
> > itk. This mesh structure for discrete surfaces is based on C-GAL and the
> > so called "Quad-Edge" structure.
> > The impact on the curent problem you're addressing here is twofold:
> > 
> > 1. This structure natively hold primal and dual connectivity information
> > of the surface, making the actual simplex related classes kind of
> > obsolete. I would like to volunteer to recode the existing classes
> > (well, triangle to simplex and simplex to triangle are not necessary
> > anymore) using the new structure and I was wondering if you would mind
> > giving me some e-mail assistance if I have any problem? I plan to do
> > that within the following 30 days.
> > 
> > 2. We are planing to port most of the surface processing filters from
> > vtk to itk to be able to have all the surface processing done in itk,
> > and only the vizualization in vtk. Has we implemented the euleroperators
> > in the new structure, this work is made easier. We have already ported
> > decimate (first version, no topology modification), smoothing, and
> > implemented an improved version of connectivity filter. More are on the
> > way. 
> > 
> > cheers,
> > 
> > alex
> > 
> > On Thu, 2006-08-24 at 02:50, Leila Baghdadi wrote:
> > 
> >>Hi Zein,
> >>
> >>I have done some further investigation with this and fixed a few minor
> >>things which I will commit to cvs soon.
> >>
> >>However, the major issue which I still have not solved is as follows,
> >>
> >>As you know, this application reads a binary a image to be converted to
> >>a mesh using the vtkMarchingCubes class. This class happens to create
> >>4-5 triangles per voxel of image (i.e, a huge number of triangles)
> >>converting such a mesh to a simplex mesh for deformation will result in
> >>a disaster unless you have unlimited memory on your system. 
> >>
> >>So To avoid the above I placed the vtkDecimate class right after
> >>vtkMarchingCubes for the sole purpose of reducing the number of
> >>triangles by a huge number (i.e, creating a simplex mesh which can
> >>easily be deformed for many iteration)
> >>
> >>Unfortunately, over the course of VTK going through major modifications
> >>this class is replaced by vtkDecimatePro which according to my tests
> >>does not do a good job of reducing the number of Triangles unless
> >>PreserveTopology is Off at which case it creates holes in the mesh
> >>(i.e., not usable for converting to simplex mesh and going through the
> >>deformation)
> >>
> >>I have sent an e-mail to vtk list asking if the old class could be added
> >>to vtk tree so I could modify my itk code to use the old class.
> >>Meanwhile, I am going to have to think about this and see if I can
> >>resolve this myself.
> >>
> >>Please continue using itk,
> >>
> >>Leila 
> >>
> >>_______________________________________________
> >>Insight-users mailing list
> >>Insight-users at itk.org
> >>http://www.itk.org/mailman/listinfo/insight-users
> > 
> > 
> > _______________________________________________
> > Insight-users mailing list
> > Insight-users at itk.org
> > http://www.itk.org/mailman/listinfo/insight-users
> > 
> > 
> 



More information about the Insight-developers mailing list