I am trying to use mrmesh to obtain 3D mesh of outer skin surface of my subjects from my T1 images.
However, I am not able to visualize produced result. Is it possible to visualize the mesh in mrview?
I also tried to load the mesh to meshlab, but I was not successful neither with the .obj and .stl format. With .stl it produces the error: premature end of file, with .obj it stuck at 33% with infinite loop producing following repeating identical output to terminal:
Warning there is a degenerate polygon of 3 verteces that was triangulated into 0 triangles
(113.000000 264.000000 64.000000)
(113.000000 264.000000 64.000000)
(112.000000 264.000000 64.000000)
I am trying to use mrmesh to obtain 3D mesh of outer skin surface of my subjects from my T1 images.
Bear in mind that this command is not tailored toward any particular form of anatomical segmentation or such; it’s a straight implementation of the Marching Cubes algorithm, combined with automatic image intensity threshold determination. So although you should get a connected mesh corresponding to the outer skin surface, you’ll likely get a lot of other stuff as well. Just making sure that your expectations of what this command could do for you are only moderate…
Is it possible to visualize the mesh in mrview?
Unfortunately not. I’d really like to have a mesh visualisation tool in mrview, but unless it becomes a necessity for a particular project I’m working on, I can’t personally justify putting the time into it. It should actually be relatively easy to implement compared to some of the other tools we already have; it’s just the amount of time involved.
I also tried to load the mesh to meshlab, but I was not successful neither with the .obj and .stl format.
I just tried meshing a T1 and it imported OK into the mesh viewer I have on hand. Though the result didn’t look as smooth as it should, there’s a little bug in there that will be fixed soon. There’s an outside chance that this mistake may be related to the degenerate polygon error you quoted.
File format ‘standards’ such as .stl and .obj are often not adequately strict, and the file import / export code of different softwares have different levels of fragility. So it can be an interesting exercise trying to produce files that succeed in obeying the expectations of a range of softwares. I’ll need to install and test meshlab myself to try to figure out what’s going on there. I’ll let you know if I find anything.