SCI Seg3D Mailing List

Text archives Help


Re: [Seg3D] seg3d Isosurface module


Chronological Thread 
  • From: Jess Tate <jess@sci.utah.edu>
  • To: seg3d@sci.utah.edu
  • Subject: Re: [Seg3D] seg3d Isosurface module
  • Date: Wed, 5 Aug 2015 10:29:32 -0600
Hi Paul,

Marching cubes is  the standard isosurfacing algorithm these days.  What is nice about it is that it is quite efficient and will not interpolate or smooth the surface for you.  this is important for segmentation to see exactly what you segmented.  If you want a lower resolution mesh, there are two things in Seg3D you can do, which are very similar in mechanism.  First is to downsample your segmentation then generate the isosurface.  Alternatively, you can decrease the isosurface quality (iso button near the - and + buttons at the bottom of the group).  Changing the isosurface quality will not affect isosurfaces already generated, so you may have to destroy an and recreate an existing one.

One issue with marching cubes is that it's resolution will be dependent on the  starting mesh/grid size, and it will be more or less uniform.  If you want adaptive mesh resolution, you are looking at a meshing software, which isn't really the scope of Seg3D. We do have meshing software (Cleaver and Biomesh3D) which may give you the kind of mesh that you are looking for, but they are geared more toward Volume meshes.  SCIRun has some meshing functionality, and you could try some other software like MeshLab or gmesh.  

Does that help? feel free to ask questions on this or the scirun-users list.  

cheers,
Jess




On Aug 4, 2015, at 11:27 AM, Paul Holcomb <paul.holcomb@gmail.com> wrote:

Is there any plan to implement a more efficient isosurfacing method in Seg3D? The current method produces meshes with significantly more vertices and faces than necessary to represent the segmentation. I routinely have to reduce my model complexity by at least 90%.

Paul


On Tue, Aug 4, 2015, 1:22 PM Purbasha Garai <pg0222@gmail.com> wrote:

Thank you so much

On Aug 4, 2015 1:11 PM, "Alan Morris" <alan.morris@carma.utah.edu> wrote:
Yes, it's marching cubes.

Alan


On Tue, Aug 4, 2015 at 10:28 AM, Jess Tate <jess@sci.utah.edu> wrote:
Hi Purbasha,

It's probably a marching cubes algorithm.

cheers,
Jess




On Aug 4, 2015, at 8:06 AM, Purbasha Garai <pg0222@gmail.com> wrote:

> Hello,
>
> What algorithm does the isosurface module use? I need to add it in my dissertation.
>
>
> Regards
> Purbasha




Archive powered by MHonArc 2.6.18.

Top of page