Sean Spicer (spicer@bme.stanford.edu)
Tue, 5 Oct 1999 16:50:32 -0700 (PDT)
My guess is that you are witnessing artifacts caused by
non-viewport aligned sampling of the volume. Volumizer defaults to an
axis-aligned sampling to take advantage of 2D texture-mapping hardware
when the 3D texturing extensions are not supported (as on o2).
Essentially what happens is as follows:
1) Volumizer determines which is the primary axis-alignment (x,y,z) of
your volume based on the current viewport, modelview, and projection,
matrices.
2) Volumizer polygonizes the specified tetra-set along this axis and maps
the correct voxels in the voxel matrix to each polygon.
Now, one thing to note is that while using 2D interpolation, intra-voxel
interpolation is not possible, so the maximum sampling resolution is
limited by the voxel dimensions of your volume (eg. a 512x256x64 volume
will have at _most_ 512 samples on the x axis, 256 samples on the y, and
64 on the z).
Most likely what is happening is that when you move inside the volume, you
see the individual slices sampled along the cardinal axis, which is not
viewport aligned. You can try pre-processing the data to have larger
dimensions, and thus decrease the size of the inter-voxel empty space, but
this will also increase the overall volume size in core memory.
Hope that answers your question,
sean
On Tue, 5 Oct 1999, Zerbini Enrico - Mardessich Senko wrote:
> Hi,
> we are trying to navigate (move inside the volume visualisation).
> Our machine is a SGIO2, which does not support the Texture 3D.
> We have created a class to visualize the volumes with Volumizer,
> obtaining a
> good result if we stay out of the volume.
> We get close to the volume using glLookAt, but getting inside the volume
> we see
> a number of superemposed slices without continuity.
> Can the problem be connected to the projection matrix?
> Have you got any idea or help concerning it?
> Thank you. Flyone
>
___________________________________________________________________________
Sean Spicer Stanford University Medical Center
Biomechanical Engineering Division of Vascular Surgery, Suite H3642
Cardiovascular Biomechanics Lab Stanford CA, 94305
Telephone...650.723.1695
Fax.........650.723.8762
http://solvedeath.stanford.edu/~spicer
This archive was generated by hypermail 2.0b2 on Tue Oct 05 1999 - 19:38:09 PDT