CGAL users discussion list

[Mariette Yvinec <>]

We did something similar for arbitrary shapes in Marie Samozino's thesis
http://tel.archives-ouvertes.fr/tel-00336379/

The main idea is to sample points on the boundary of the objets
(i.e on your polyhedron)
compute the Voronoi diagram - Delaunay triangulation of these
sample points and select a subset of the poles
as centers of approximating balls.
Pole is defined as in Amenta and Bern paper.
Two poles (at most) are defined for each sample
where a pole is either
the vertex  of the Voronoi cell that is furthest  from
the cell site, or the vertex  of the Voronoi cell that is the furthest among vertices in the halfspace
opposite to the furthest.


Le 21/06/12 01:12, Zohar a écrit :

Hi,

I'm looking for a way to roughly fill a polyhedron with minimum/largest
spheres (but again, roughly). One way is to approximate the medial axis and
distribute spheres along it. Any ideas how to accomplish that?

Thanks

--
View this message in context: http://cgal-discuss.949826.n4.nabble.com/Fill-a-polyhedron-with-spheres-tp4655291.html
Sent from the cgal-discuss mailing list archive at Nabble.com.

-- 
Mariette Yvinec
Geometrica project team
INRIA  Sophia-Antipolis