CGAL users discussion list

[Bruno Manganelli <>]

I think there is a missing return statement in make_skin_suface_mesh_3.h after line 37.

Adding that fixed my issue.

On Mon, Apr 26, 2021 at 11:21 AM Sebastien Loriot <> wrote:

Do you have a minimal example we could use to reproduce the issue?

Did you also try this example:
https://doc.cgal.org/latest/Skin_surface_3/Skin_surface_3_2union_of_balls_simple_8cpp-example.html

Best,

Sebastien.

On 4/26/21 11:09 AM, Bruno Manganelli ( via
cgal-discuss Mailing List) wrote:
> That indeed is exactly what I needed, thanks!
> Without mesh subdivision and shrink_factor of 1, the algorithm takes
> less than 5s.
> However, trying to execute CGAL::subdivide_skin_surface_mesh_3 on the
> surface generated with the shrink_factor of 1 results in an assertion
> failure in in triangulate_mixed_complex_3.h:983
> CGAL_assertion(orientation(ch) == POSITIVE);
> Is this expected? I couldn't find the requirement that shrink_factor
> must be < 1 in the documentation.
>
> On Mon, Apr 26, 2021 at 8:26 AM Sebastien Loriot <
> <mailto:>> wrote:
>
>     Andreas mentioned that there is the following package that also might
>     fit you need (in case you want more that spheres):
>
>     https://doc.cgal.org/latest/Skin_surface_3/index.html#Chapter_3D_Skin_Surface_Meshing
>     <https://doc.cgal.org/latest/Skin_surface_3/index.html#Chapter_3D_Skin_Surface_Meshing>
>
>     Best,
>
>     Sebastien.
>
>     On 4/25/21 11:42 AM, Sebastien Loriot wrote:
>      > Did you switch on parallel meshing?
>      > By default the points on the spheres are computing by bisection.
>      > It is possible to provide a specific functor for computing points
>      > on the surface.
>      >
>      > Best,
>      >
>      > Sebastien.
>      >
>      > On 4/25/21 11:12 AM, Bruno Manganelli (
>     <mailto:> via
>      > cgal-discuss Mailing List) wrote:
>      >> Thanks for all your dedication to help :)
>      >> I have tried your method, and indeed it works. However it is still
>      >> fairly slow, taking around 70 s to process 761 spheres (on a
>     ryzen 5
>      >> 3600). Do you think there are any optimizations that I could try
>     or do
>      >> you think it will be hard to significantly speed  up this approach?
>      >>
>      >>
>      >> On Sat, Apr 24, 2021 at 2:50 PM Sebastien Loriot
>      >> < <mailto:>
>     <mailto: <mailto:>>> wrote:
>      >>
>      >>     With the following code (quick and dirty):
>      >> https://gist.github.com/sloriot/d563b5ce27d28fcfd73beb1b628d72b5
>     <https://gist.github.com/sloriot/d563b5ce27d28fcfd73beb1b628d72b5>
>      >>   
>     <https://gist.github.com/sloriot/d563b5ce27d28fcfd73beb1b628d72b5
>     <https://gist.github.com/sloriot/d563b5ce27d28fcfd73beb1b628d72b5>>
>      >>
>      >>     I could generate a mesh.
>      >>
>      >>     Best,
>      >>
>      >>     Sebastien.
>      >>
>      >>     PS: there might be some initialization issues if you have
>     several
>      >>     (small) connected components and initial points should be added.
>      >>     If you need clean intersections then some extra work is also
>     needed.
>      >>
>      >>
>      >>     On 4/24/21 2:06 PM, Sebastien Loriot wrote:
>      >>      > If you take the maximum radius, you can collect all the
>     closest
>      >> point
>      >>      > to the query and stop when you exceed the maximum radius.
>      >>      > You can associate the radius to each point like in this
>     exemple:
>      >>      >
>      >>      >
>      >>
>      >>
>     https://doc.cgal.org/latest/Spatial_searching/Spatial_searching_2searching_with_point_with_info_8cpp-example.html
>     <https://doc.cgal.org/latest/Spatial_searching/Spatial_searching_2searching_with_point_with_info_8cpp-example.html>
>
>      >>
>      >>
>      >>
>     <https://doc.cgal.org/latest/Spatial_searching/Spatial_searching_2searching_with_point_with_info_8cpp-example.html
>     <https://doc.cgal.org/latest/Spatial_searching/Spatial_searching_2searching_with_point_with_info_8cpp-example.html>>
>
>      >>
>      >>
>      >>      >
>      >>      >
>      >>      > HTH,
>      >>      >
>      >>      > Sebastien.
>      >>      >
>      >>      > On 4/24/21 11:28 AM, Bruno Manganelli
>     ( <mailto:>
>      >>     <mailto:
>     <mailto:>> via
>      >>      > cgal-discuss Mailing List) wrote:
>      >>      >> Thanks for the reply Sebastien :)
>      >>      >> Unfortunately I can't use a kd-tree because the spheres
>     can have
>      >>      >> different radii, so a point won't necessarily be inside the
>      >> closest
>      >>      >> sphere.
>      >>      >> I have tried adapting an AABB tree to use spheres as
>     primitives,
>      >>     but
>      >>      >> it doesn't seem to work when querying for intersections
>     with
>      >> points.
>      >>      >> I admittedly don't know how the bounding boxes are
>     created or
>      >>     whether
>      >>      >> querying for intersections is a valid thing to do.
>      >>      >>
>      >>      >> For reference, this is what I was doing with AABB trees:
>      >>      >>
>      >>      >> class AABBSpherePrimitive {
>      >>      >>   public:
>      >>      >>     using Point = Point_3;
>      >>      >>     using Datum = Sphere_3;
>      >>      >>     using Point_reference = Point;
>      >>      >>     using Datum_reference = Datum;
>      >>      >>     using Id = Datum;
>      >>      >>
>      >>      >>     AABBSpherePrimitive(Sphere_3 sphere)
>      >>      >>         : m_sphere_datum(sphere)
>      >>      >>     {
>      >>      >>       m_reference_point = sphere.center();
>      >>      >>       m_reference_point +=
>      >>      >> K::Vector_3{std::sqrt(sphere.squared_radius()), 0, 0};
>      >>      >>     }
>      >>      >>
>      >>      >>     Datum_reference datum() const { return m_sphere_datum; }
>      >>      >>     Id id() const { return m_sphere_datum; }
>      >>      >>     Point_reference reference_point() const { return
>      >>     m_reference_point; }
>      >>      >>   private:
>      >>      >>     Datum m_sphere_datum;
>      >>      >>     Point m_reference_point;
>      >>      >> };
>      >>      >>
>      >>      >> using SphereTreeTraits = CGAL::AABB_traits<K,
>      >> AABBSpherePrimitive>;
>      >>      >> using SphereTree = CGAL::AABB_tree<SphereTreeTraits>;
>      >>      >>
>      >>      >>
>      >>      >> On Sat, Apr 24, 2021 at 7:42 AM Sebastien Loriot
>      >>      >> < <mailto:>
>     <mailto: <mailto:>>
>      >>     <mailto: <mailto:>
>     <mailto: <mailto:>>>> wrote:
>      >>      >>
>      >>      >>     What I would do is generate a mesh for the union of
>      >> spheres by
>      >>      >> adapting
>      >>      >>     this example:
>      >>      >>
>      >>      >>
>      >>      >>
>      >>
>      >>
>     https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html>
>
>      >>
>      >>
>      >>
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html>>
>
>      >>
>      >>
>      >>      >>
>      >>      >>
>      >>      >>
>      >>
>      >>
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html>
>
>      >>
>      >>
>      >>
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html
>     <https://doc.cgal.org/latest/Mesh_3/Mesh_3_2mesh_implicit_sphere_8cpp-example.html>>>
>
>      >>
>      >>
>      >>      >>
>      >>      >>
>      >>      >>     to handle several spheres.
>      >>      >>
>      >>      >>     To do that, I think the simplest way to do it is to
>     use a
>      >>     kd-tree
>      >>      >> [1] to
>      >>      >>     get the closest sphere center from the query point
>     and then
>      >>     use the
>      >>      >>     existing formula with the selected sphere.
>      >>      >>
>      >>      >>     Let us know if you could do it.
>      >>      >>
>      >>      >>     Best,
>      >>      >>
>      >>      >>     Sebastien.
>      >>      >>
>      >>      >>     [1]
>     https://doc.cgal.org/latest/Spatial_searching/index.html
>     <https://doc.cgal.org/latest/Spatial_searching/index.html>
>      >>     <https://doc.cgal.org/latest/Spatial_searching/index.html
>     <https://doc.cgal.org/latest/Spatial_searching/index.html>>
>      >>      >>   
>     <https://doc.cgal.org/latest/Spatial_searching/index.html
>     <https://doc.cgal.org/latest/Spatial_searching/index.html>
>      >>     <https://doc.cgal.org/latest/Spatial_searching/index.html
>     <https://doc.cgal.org/latest/Spatial_searching/index.html>>>
>      >>      >>
>      >>      >>
>      >>      >>
>      >>      >>     On 4/23/21 11:57 PM, bmanga (
>     <mailto:>
>      >>     <mailto:
>     <mailto:>>
>      >>      >>     <mailto:
>     <mailto:>
>      >>     <mailto:
>     <mailto:>>> via cgal-discuss
>      >>      >>     Mailing List) wrote:
>      >>      >>      > I have also tried to use
>      >>      >>      >
>      >>     PMP::experimental::autorefine_and_remove_self_intersections, but
>      >>      >>     it fails
>      >>      >>      > with the following:
>      >>      >>      >
>      >>      >>      > CGAL error: precondition violation!
>      >>      >>      > _expression_ : vaa != vbb
>      >>      >>      >
>      >>      >>      >
>      >>      >>      >
>      >>      >>      > --
>      >>      >>      > Sent from:
>     http://cgal-discuss.949826.n4.nabble.com/
>     <http://cgal-discuss.949826.n4.nabble.com/>
>      >>     <http://cgal-discuss.949826.n4.nabble.com/
>     <http://cgal-discuss.949826.n4.nabble.com/>>
>      >>      >>     <http://cgal-discuss.949826.n4.nabble.com/
>     <http://cgal-discuss.949826.n4.nabble.com/>
>      >>     <http://cgal-discuss.949826.n4.nabble.com/
>     <http://cgal-discuss.949826.n4.nabble.com/>>>
>      >>      >>      >
>      >>      >>
>      >>      >>     --     You are currently subscribed to cgal-discuss.
>      >>      >>     To unsubscribe or access the archives, go to
>      >>      >> https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>
>      >>     <https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>>
>      >>      >>     <https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>
>      >>     <https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>>>
>      >>      >>
>      >>      >>
>      >>      >>
>      >>      >> --
>      >>      >> You are currently subscribed to cgal-discuss.
>      >>      >> To unsubscribe or access the archives, go to
>      >>      >> https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>
>      >>     <https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>>
>      >>      >>
>      >>
>      >>     --     You are currently subscribed to cgal-discuss.
>      >>     To unsubscribe or access the archives, go to
>      >> https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>
>      >>     <https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>>
>      >>
>      >>
>      >>
>      >> --
>      >> You are currently subscribed to cgal-discuss.
>      >> To unsubscribe or access the archives, go to
>      >> https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>
>      >>
>
>     --
>     You are currently subscribed to cgal-discuss.
>     To unsubscribe or access the archives, go to
>     https://sympa.inria.fr/sympa/info/cgal-discuss
>     <https://sympa.inria.fr/sympa/info/cgal-discuss>
>
>
>
> --
> You are currently subscribed to cgal-discuss.
> To unsubscribe or access the archives, go to
> https://sympa.inria.fr/sympa/info/cgal-discuss
>

--
You are currently subscribed to cgal-discuss.
To unsubscribe or access the archives, go to
https://sympa.inria.fr/sympa/info/cgal-discuss