vtkMarchingCubes and vtkImageMarchingCubes fail on large Datasets
This issue was created automatically from an original Mantis Issue. Further discussion may take place here.
using both of these classes can result in a blank screen if the dataset is too big and the threshold value produces alot of "hits" (e.g. CT-scan with threashold around 1000 (bones)). It is not a problem of the amount of Ram, since swap file was set to extreme large value (10GB) for testing.
pipeline is very simple:
// vtkVolume16Reader v16->SetDataDimensions (lVolumeInfo.XDim,lVolumeInfo.YDim); v16->SetImageRange (0,lVolumeInfo.ZDim); v16->SetDataByteOrderToLittleEndian(); v16->SetFilePrefix(filename.c_str()); v16->SetDataSpacing (lVolumeInfo.XSpacing, lVolumeInfo.YSpacing, lVolumeInfo.ZSpacing); v16->ReleaseDataFlagOff(); v16->Update();
// marching cubes iso_Extractor->SetInputConnection(median3D->GetOutputPort()); iso_Extractor->SetValue(0, 1068); iso_Extractor->ComputeGradientsOn(); iso_Extractor->ComputeNormalsOn(); iso_Extractor->ReleaseDataFlagOff();
// the mapper for the iso_extractor iso_Mapper->SetInput(iso_Extractor->GetOutput()); iso_Mapper->ScalarVisibilityOff(); iso_Mapper->ReleaseDataFlagOff();
// the actor for the iso_extractor iso_Actor->SetMapper(iso_Mapper); iso_Actor->SetPosition(0,0,0);
DataSet is 431 slices CT-Data with 512px. (works with oder classes without problems). This behaviour happens on most big DataSets.
Using vtkImageMarchingCubes with Streaming didn't help either.