CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

One way to do it would be to build an AABB-tree of the triangles of the
input (or make a copy of the mesh and do an AABB-tree of its faces).

Then you can plug into your cost functor a distance query of the 
placement to the AABB-tree (don't forget to call 
acceleterate_distance_queries()).
If the placement is further away from your threshold then return an
empty cost. This will not give you a certified distance (like Hausdorff
would do) but at least you'll have the guarantee that all the mesh
points are within a given distance range from the input.

Best,

Sebastien.



On 8/2/19 10:58 PM, Audette, Michel A. wrote:
> Dear CGAL experts,
>
>
> I would like to use the Triangulated Surface Mesh Simplification 
> algorithm to modify a triangulated model of the skull base, in a manner 
> that keeps the evolving skull surface in close proximity to a set of 
> triangulated surfaces of the cranial holes (foramina). In other words, 
> if we don't impose this type of constraint, the decimation tends to make 
> these holes arbitrarily big, which is the opposite of what we want. Is 
> there a simple way to use this code, or somehow modify it, to consider 
> the faces of a second surface B (holes) as a constraint for the 
> simplification of surface A (cranial bone)?
>
>
> Thanks for your kind support.
>
>
> Michel Audette, Ph.D.
> Associate Professor,
> Department of Modeling, Simulation and Visualization Engineering,
> Old Dominion University,
> Norfolk, VA.