CGAL users discussion list

[Sebastien Loriot <>]

If you do not allow the addition of new points nor move of existing
points, then there is no way a sliver could be removed from Delaunay
as it is uniquely defined given a set of points.

What are you allowed to do with your point set?

Best,

Sebastien.

On 3/11/24 11:48, Justice Temiloluwa Olatunbosun wrote:
> Dear Sir/Madam,
>
> I am Justice Olatunbosun, a master student of mechanical engineering and 
> a scientific assistant of computational mechanics at the Dortmund 
> Technical University, Germany.
>
> I am currently using the CGAL library in research areas applied to 
> generating finite element meshes. Currently, most of my work has been in 
> two dimensional space with the CGAL 2D delauanay tessellations with much 
> success. By demand, I am required to make extensions to three 
> dimensional space as well. However, the CGAL 3D delauanay triangulation 
> is not sufficient alone in itself to generate usable finite element 
> meshes for my work due to the presence of slivers.
>
> CGAL, offers the option to refine and optimize mesh which removes 
> slivers, but this option is only available to the 
> "CGAL::Mesh_triangulation_3::type" which requires a domain definition in 
> its creation. For my work, I only intend to provide vertex coordinate of 
> the Point_3 as input and as such, only the "Delaunay_triangulation_3.h" 
> works for me. I am currently unable to find sliver treatment for direct 
> CGAL 3D triangulation in the documentation manual. I would like to find 
> out if this option is available somewhere via CGAL or what other methods 
> are available to generate meshes from only coordinates of vertices and 
> then refined against slivers.
>
> I would appreciate your kind feedback.
>
> Best Regards
> Justice Olatunbosun