[Paraview] Bad triangulation for Polygon

Mike Jackson imikejackson at gmail.com
Tue Oct 2 12:02:43 EDT 2007


I have had the same problems using the triangulation routines from  
ParaView/VTK. I ended up using some Delaunay Triangulation Code from  
the "Quake" Project at CMU and turning that code into a ParaView  
Filter (or at least trying to). I have had decent success with this  
approach. The filter works for most cases I throw at it but there are  
some cases where my algorithm is not properly preparing the input  
dataset.

http://www.cs.cmu.edu/~quake/triangle.html

http://cm.bell-labs.com/netlib/voronoi/triangle.zip

So, basically, you are probably on your own for writing/implementing/ 
borrowing better triangulation code.

-- 
Mike Jackson   Senior Research Engineer
Innovative Management & Technology Services


On Oct 2, 2007, at 11:49 AM, Bob Anderson wrote:

> For polygons with lobes, the tests I’ve done with ParaView don’t  
> give correct surface triangulation.
>
> In the case where the object comes from my custom reader, using  
> output->SetPolys(myCellsPtr); it produces a surface that doesn’t  
> stay within the polygon boundary, and applying the Delaunay2D  
> filter to the polyline object (SamplePolygon1Cell.vtk) also  
> ‘blobbs’ the polygon.
>
> See attachments –a correct surface triangulation  
> (SamplePolySurface.jpg), and the ParaView result  
> (PV_Delaunay2D_Polygon.jpg)
>
> I’m currently doing my own triangulation using ‘hgrd_src’ routines  
> from CodeProject, but if there is a better way to get the correct  
> results within ParaView I’d be happy to try it.
>
>
> Bob Anderson
> MIRARCO Mining Innovation
> Laurentian University
>
> <PV_Delaunay2D_Polygon.jpg>
> <SamplePolygon1Cell.vtk>
> <SamplePolySurface.jpg>
> _______________________________________________
> ParaView mailing list
> ParaView at paraview.org
> http://www.paraview.org/mailman/listinfo/paraview

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/paraview/attachments/20071002/b1927d06/attachment.htm


More information about the ParaView mailing list