[vtkusers] Re: Please Help : 3D reconstitution and contour Filtring

Michael Knopke Michael.Knopke at gmx.de
Wed Feb 7 10:17:18 EST 2007


Hi Mohamed,

sorry to say, but I'm facing the same problems here.
I think it's an memory error, because it depends on the size of the volume
and the number of "hits" the marching cube algorithm will have. For example,
you can use vtkImageShrink3D on you dataset to reduce the number of polygons
you will create. This will help showing at least an 3D model (not always
though). It also depends what value (eg. density) you are trying to find in
an dataset. For example, if you are looking for around 1000 (bone) in Ct, it
will have a high number of hits and therefor more often fail. The cause for
this should not be the amount of RAM, since Windows will use Paging (i tried
to set up the pagefile to an riddiculous amount, so no help here).
I put this into the bug tracking database, but no response yet (can be
found on page 11 (vtk)). I also tried vtkImageMarchingCube which has
streaming capabilities but no help from this either and this class produced
alot of severe access violations with certain values, so I don't recommend it.
Anyway, I might do something wrong or if anybody knows how to querry for this memory limitation before rendering into nirvana would help me alot.

Regarding the time it takes: You will reduce the calculation time to a maximum if you reduce the output beforehand with vtkImageShrink3D, but with a cost of quality of course.

Regards

Michael
-- 
Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! 
Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer



More information about the vtkusers mailing list