[Insight-users] More information about itkBinaryMask3DMeshSource: Marching Cubes options

[Luis Ibanez]

Hi Samuel,

Yeap, the algorithm in the BinaryMask3DMeshSource is
basically Marching Cubes.

Note that this algorithm is patented.

It is probably the only patented unit in ITK, so you
may want to reconsider it use if you plan to develop
a commercial application.

As an alternative for visualizing the results of a
segmentation you could also go through converting
the segmented image to VTK using ImageToVTKImage
and then do the classical processing in VTK that
Felix just mentioned. (notice that the Marching
cubes in VTK is also pattented...)

Here is a list of options for performing

           ImageMask to Mesh

with the aim of Visualizating segmented images.



-------------------------------------------

1) itkBinaryMask3DMeshSource

http://www.itk.org/Insight/Doxygen/html/classitk_1_1BinaryMask3DMeshSource.html

algorithm: Marhching Cubes
status:    Patented: free for non-commercial use
input:     ITK image
output:    ITK mesh
note:      Need to convert the output ITK mesh
            to a VTK poly data (if you want to
            visualize with VTK).

------------------------------------------

2) vtkImageMarchingCubes

http://www.vtk.org/doc/nightly/html/classvtkImageMarchingCubes.html

algorithm: Marching Cubes
status:    Patented: free for non-commercial use
input:     VTK Image Data
output:    VTK Poly Data
note:      need to convert ITK image to VTK
            image in order to plug the input

------------------------------------------

3) vtkSynchronizedTemplates

http://www.vtk.org/doc/nightly/html/classvtkSynchronizedTemplates3D.html

algorithm:  Synchronized Templates
status:     Patent pending: free for non-commercial use
input:      VTK Image Data
output:     VTK Poly Data
note:       Need to convert the ITK image into
             a VTK Image in order to plug the filter
             in a pipeline.
             This algorithm is faster than Marching
             cubes, by a factor or 3 to 4.

------------------------------------------

4) vtkContourImageFilter

http://www.vtk.org/doc/nightly/html/classvtkContourFilter.html

algorithm:  Extract cell by cell & compute the N-1 D cell
status:     non-patented
input:      VTK Data Set
output:     VTK Poly Data
note:       Performance is lower due to the generality of
             the algorithm used. Probably about 10 times
             slower than marching cubes.


--------------------------------------------



Please let us know if you have further questions.



      Luis


-------------------------------------------
Samuel Rodríguez Bescos wrote:
> Hello every body,
>  
> Does anybody know what is the algorithm used in the filter 
> itkBinaryMAsk3DMeshSource?. I think that is the marching Cubes explained 
> in "The visualization Tookit" Book of Schoeder, Martin and Lorensen but 
> I'm not sure.
>  
> Could anybody tell me if It's the same or if there is any difference?
>  
> Thanks in advance.
>  
> Sam