[vtkusers] variable density grid contour algorithm

[Bjorn Oksnevad]

Hello,
I am struggling putting together the right mix of vtk pipeline components to
get the the effect I am looking for.  I'm hoping someone can help me out.

I would like to visualize an experiment, where sample data is measured at
various points in 3d space.  The sampling points are regular in topology,
and cylindrical in geometry.  However, some areas in 3d space are sampled
with a higher density than others.  I would like to visualize a contour
surface of the data for a given measurement value (scalar value).

I have tried using vtkStructuredGrid with blank points to handle the varying
density of data.  This worked well, except that my contour surface had holes
where data was missing.  Is there a vtk filter that fills in holes to a
contour surface?  Or a vtk filter that fills in blank data points in a
vtkStructuredGrid?

I also briefly played around with representing the data as
vtkUnstructuredGrid and vtkPolyData.  I wasn't successful getting a contour
surface with these approaches.

Any high level vtk pipeline tips for achieving my desired effect?  Is
structured grid the right approach, or am I better off starting with
unstructured or poly data?

Thanks!
Bjorn
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.vtk.org/pipermail/vtkusers/attachments/20110321/bed63290/attachment.htm>