Differences in paraview and vtk volume rendering?

I am trying to create a volume rendering in vtk that resembles the paraview output as closely as possible.

  1. Is it possible to produce the same volume rendering with paraview and vtk?
  2. Are there known settings or pipeline changes I need to make in vtk in order to accomplish this?

Example -
Test data set - vtkImageData - Scalar point data - single component scalar with a range of about 0 to 0.4
testdata.vti (3.8 MB)

Basic setup - Load in data. Show as a volume representation. Set color to constant red. Set opacity function as ramp that goes from 0 to 1.


VTK (shown here using vtkplotter)
I had to set the mode=1 (maximum intensity) for the volume rendering to show anything. The default is composite blending mode, which is the same mode used in all/most of the VTK example themselves. In this case, the composite blending mode results in a barely visible volume rendering. Iā€™m not sure why that is. I have also attempted all this with C++ using the smart volume mapper with similar results.

import vtk
from vtkplotter import datadir, load, Volume, show

reader = vtk.vtkXMLImageDataReader()
img = reader.GetOutput()
vol1 = Volume(img, mode=1) 
vol1.color(["red", "red"])
vol1.alpha([0, 1])
show(vol1, "vtk plotter", axes=1, resetcam=True, bg='black')


Thank you

Figured this out. The fix in my case was to call SetScalarOpacityUnitDistance(0.01) on the vtkVolumeProperty. This was roughly the value being used by paraview by default. This also allows me to use the composite blending mode and get a good volume rendering.

1 Like