[Insight-developers] Surface Class derived from Mesh

[Stephen R. Aylward]

Hi,

Andrew is out of town until Tuesday.   

To answer your question...his proposed Surface class derives from mesh -
it just simplifies/specializes mesh's API for handling surfaces.   Then
just as there are filter as well as point-by-point interpolation
methods, we'd have surface approximation/interpolation methods as you
suggest.

s

"Miller, James V (CRD)" wrote:
> 
> Andrew,
> 
> Can we talk about this at one of the tcons?  It's not clear how this is going to be different from
> mesh.  Mesh allows you to wire up your points anyway you would like to form a surface.
> 
> Are you referring to different algorithms to "reconstructing" surfaces from a point cloud (i.e.
> generating topology from a set of points)?  If so, I would not create a new surface class but would
> write a series of point set to mesh filters. Each algorithm would be a different filter (Delaunay,
> Hoppe's method, etc.)
> 
> -----Original Message-----
> From: Andrew MacKelfresh [mailto:mackelfr@cs.unc.edu]
> Sent: Tuesday, August 07, 2001 3:26 PM
> Cc: insight-Developers
> Subject: [Insight-developers] Surface Class derived from Mesh
> 
> Hi all -
> 
> I am working on a project that specifies a three dimensional surface -
> this surface is made up of a set of points.
> 
> I would like to create a class to represent these points and propose
> that it be structured as follows:
> 
> 1)  It would be derived from the Mesh class because a surface is
> essentially a Mesh.
> 
> 2)  It would allow for the addition of points and extraction of points -
> this functionality would stem from the superclass of Mesh, PointSet.
> 
> 3)  What would make Surface unique is that there would exist
> functionality within it to specify the connectivity of the points of the
> Surface.  A default connectivity function would exist within Surface and
> if you wanted to specify a different connectivity, you could derive the
> Surface class and specify your own connectivity algorithm.
> 
> Is anyone doing something similar to this?  If not, I propose to do this
> and would appreciate feedback.
> 
> Thanks,
> Andrew
> 
> _______________________________________________
> Insight-developers mailing list
> Insight-developers@public.kitware.com
> http://public.kitware.com/mailman/listinfo/insight-developers
> 
> _______________________________________________
> Insight-developers mailing list
> Insight-developers@public.kitware.com
> http://public.kitware.com/mailman/listinfo/insight-developers

-- 
=======================================================================
Stephen R. Aylward, Ph.D.			www.cs.unc.edu/~aylward
Research Assistant Professor of Radiology		aylward@unc.edu
Adjunct Assistant Professor of Computer Science		  (919)966-9695