CGAL users discussion list

["Grimm, Raphael (IAR)" <>]

Hi,
Thanks for the idea, but the problem with a distance query is:
* it is slow (since it extract more information, than the binary label I
need for my application) 
* after the query, I do not know whether the point is inside or outside of
the mesh (so I need to perform the inside check anyways)

To check how much slower a distance query is, I wrote some test code (it is
attached).

If I run it with mesh (21965 vertices, 43926 faces, roughly a cube), then
the tree version takes much longer (x200+ times) than the check whether the
point is inside of the mesh.
This test has to be very fast, since I have a lot of query points.

Hence I need some way to inflate my mesh.

Maybe you or someone else has another idea.

Best regards,

Raphael.

-----Original Message-----
From:  <> On
Behalf Of "Sebastien Loriot (GeometryFactory)"
Sent: Wednesday, 23 September, 2020 18:36
To: 
Subject: Re: [cgal-discuss] minkowski_sum_3 segfault

Did you try the AABB-tree and the closest_point() method?

https://doc.cgal.org/latest/AABB_tree/#title6

Best regards,

Sebastien.

On 9/23/20 4:17 PM, "Grimm, Raphael (IAR)" ( via
cgal-discuss Mailing List) wrote:
> Hello,
> 
> I have a non-convex mesh and want to check whether a point is within a 
> certain distance of it. Since distance queries are slow, I decided to 
> inflate the mesh and check whether the point is inside.
> 
> Scaling does not work, since my mesh is concave.
> 
> I did some looking around and came across minkowski_sum_3. I am 
> planning to use a sphere to inflate my original mesh by using
minkowski_sum_3.
> 
> I wrote some simple test for minkowski_sum_3 that adds a tetrahedron 
> to a cube (code is attached).
> 
> This test crashes with a segfault. Here is the stack trace:
> 
> 1  CGAL::SNC_FM_decorator<CGAL::SNC_structure<CGAL::Epick,
> CGAL::SNC_indexed_items,
> bool>>::determine_facet

> stl_vector.h                           805  0x55555559244a
> 
> 2  CGAL::SNC_FM_decorator<CGAL::SNC_structure<CGAL::Epick,
> CGAL::SNC_indexed_items,
> bool>>::create_facet_objects

> In_place_list.h                        84   0x5555555a0dea
> 
> 3  CGAL::SNC_external_structure<CGAL::SNC_indexed_items,
> CGAL::SNC_structure<CGAL::Epick, CGAL::SNC_indexed_items,
> bool>>::categorize_facet_cycles_and_create_facets
> stl_list.h                             953  0x5555555a16d7
> 
> 4  CGAL::SNC_external_structure<CGAL::SNC_indexed_items,
> CGAL::SNC_structure<CGAL::Epick, CGAL::SNC_indexed_items,
> bool>>::build_external_structure                  
> In_place_list.h                        310  0x5555555c3417
> 
> 5  CGAL::Nef_polyhedron_3<CGAL::Epick, CGAL::SNC_indexed_items,
> bool>::build_external_structure

> SNC_const_decorator.h                  125  0x5555555f6188
> 
> 6  CGAL::Nef_polyhedron_3<CGAL::Epick, CGAL::SNC_indexed_items,
> bool>::delegate

> Nef_polyhedron_3.h                     1071 0x5555555f6188
> 
> 7
> CGAL::bipartite_nary_union_sorted_combined<CGAL::Nef_polyhedron_3<CGAL
> ::Epick,
> CGAL::SNC_indexed_items,
> bool>>                                                    
> bipartite_nary_union_sorted_combined.h 214  0x5555555f6188
> 
> 8  CGAL::minkowski_sum_3<CGAL::Nef_polyhedron_3<CGAL::Epick, 
> CGAL::SNC_indexed_items, bool>>                                
>
                                         In_place_list.h                    
    
> 309  0x5555555f7384
> 
> 9  
> main

>            minkowski_sum_3.cpp                    81   0x55555555d8df
> 
> 10 
> __libc_start_main

> libc-start.c                           310  0x7ffff69ccb97
> 
> 11 _start
> 
> I have two questions:
> 
> 1.Is there a better (more efficient) way to do this?
> 
> 2.If not, how can I fix the segfault / what am I doing wrong?
> 
> I appreciate any hints.
> 
> Best
> 
> Raphael
> 
> ----
> 
> Karlsruhe Institute of Technology (KIT)
> 
> Institute for Anthropomatics and Robotics (IAR)
> 
> High Performance Humanoid Technologies (H2T)
> 
> Raphael Grimm (M.Sc.)
> 
> Research Scientist
> 
> Adenauerring 2
> 
> Building 50.20, Room 334
> 
> 76131 Karlsruhe, Germany
> 
> Phone: +49 721 608-47133
> 
> Fax: +49 721 608-48270
> 
> Email: 
> 
> Web: https://www.humanoids.kit.edu
> 
> KIT - The Research University in the Helmholtz Association
> 

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

Attachment: mesh_distance_calc_test.cpp
Description: Binary data

Attachment: smime.p7s
Description: S/MIME cryptographic signature