TriRep class - edge constraints 3D Delaunay Triangulation

Hello everyone,

I have a 3D TriRep triangualtion imported within Matlab from an .stl file and created using:

tr = TriRep(f,v(:,1),v(:,2),v(:,3)); 
% original triangulation in face-vertex format f & v

I want to use a constrained DelaunayTri to perform a spatial search within my triangulation. As indicated in the spatial searching help, I need to use the previous triangulation tr to find all edges in the triangulation using:

Cedges = edges(tr); % does it work for 3D triangulation?

and then I was thinking to generate the Delaunay Triangulation simply using the following function:

dt = DelaunayTri(v,Cedges);

Is there any possibility to generate an edge constrained DelaunayTri sub-class in 3D from a TriRep? I do not want to implement the brute force that was proposed as a workaround for spatial searching since I have a high-number of triangles and need a faster approach.

Thank you for your help,

Romain

[Romain W]

the reason is that I would like to use the Matlab built-in routines (methods) peculiar to the DelaunayTri sub-class

[Romain W]

and I am only interested in the surface mesh

[Anoop]

@Romain W, Were you able to find the solution for your problem. I am also stuck in the same situation and looking for a solution.

Thanks

Answer 1

[DGM]

The answer is no. Creating the DelaunayTri object creates a new triangulation, so whatever is defined by the faces of TR is ignored. Constrained triangulation is only supported when V is 2D. Without constraints in 3D, DT is a tetrahedral mesh filling the convex hull of the point cloud defined by V. It's probably an unhelpful blob.