vtkResampleToImage for volume rendering of big data

Support
1

Hi everyone,

I am trying to do a volume rendering of a relatively big cfd case (about 60 mln cells).
Before coding everything in c++, I made some test runs with Paraview and because of the size of the mesh, the only way to have a usable volume rendering was to Resample to Image and then display the volume.

Within vtk I am struggling to do the same, I can load my results in a vtkRectilinearGrid, vtkStructuredGrid or vtkUnstructuredGrid but afterwards I am struggling to use vtkResampleToImage and visualize it.

    vtkNew<vtkResampleToImage> resample;
    resample->SetSamplingDimensions(nx, ny, nz);  
    resample->AddInputDataObject(rectilinearGrid);
    resample->Update();

I get here, but after that I am having a bit of difficulties to render the output of resample.

I couldn’t find anything similar in the c++ examples, does anybody know what would be the proper way to do it?

2

Hello,

vtkResampleToImage outputs a vtkImageData, so you need something like this:

#include <vtkSmartVolumeMapper.h>
#include <vtkVolumeProperty.h>
#include <vtkColorTransferFunction.h>
#include <vtkPiecewiseFunction.h>


    // Create mapper (visualization parameters)
    vtkSmartPointer<vtkSmartVolumeMapper> volumeMapper = vtkSmartPointer<vtkSmartVolumeMapper>::New();
    volumeMapper->SetBlendModeToComposite(); // composite first
    volumeMapper->SetInputData( resample->GetOutput() );
    //volumeMapper->SetRequestedRenderModeToRayCast();
    volumeMapper->Update();

    //create a color transfer function
    //create a color interpolator object
    vtkSmartPointer<vtkColorTransferFunction> ctf = vtkSmartPointer<vtkColorTransferFunction>::New();
    ctf->SetColorSpaceToRGB();
    double delta = max - min;
    ctf->AddRGBPoint(min               , 0.000, 0.000, 1.000);
    ctf->AddRGBPoint(min + delta * 0.25, 0.000, 1.000, 1.000);
    ctf->AddRGBPoint(min + delta * 0.50, 0.000, 1.000, 0.000);
    ctf->AddRGBPoint(min + delta * 0.75, 1.000, 1.000, 0.000);
    ctf->AddRGBPoint(max               , 1.000, 0.000, 0.000);

    //create transfer mapping scalar value to opacity so unvalued cells are invisible
    //if the user set a no-data value.
    vtkSmartPointer<vtkPiecewiseFunction> otf = vtkSmartPointer<vtkPiecewiseFunction>::New();
    otf->AddPoint( no_data_value, 0.0 );
    otf->AddPoint( min, 1.0 );
    otf->AddPoint( max, 1.0 );

    //this object instructs VTK on how to render the volume.
    vtkSmartPointer<vtkVolumeProperty> volumeProperty = vtkSmartPointer<vtkVolumeProperty>::New();
    volumeProperty->ShadeOff();
    volumeProperty->SetColor(ctf);
    volumeProperty->SetScalarOpacity(otf);
    //volumeProperty->SetInterpolationType(VTK_LINEAR_INTERPOLATION);
    volumeProperty->SetInterpolationType(VTK_NEAREST_INTERPOLATION);

    // Finally, pass everything to the vtkVolume (some kind of actor) and pass it to the renderer.
    vtkSmartPointer<vtkVolume> volume = vtkSmartPointer<vtkVolume>::New();
    volume->SetMapper( volumeMapper );
    volume->SetProperty( volumeProperty );

    renderer->AddActor( volume );

Of course this is not the only way to render a volume, but I hope this helps to start up.

best,

PC

1 Like