CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

Again, nothing catches my eye. If you modified a demo, push it somewhere 
in github or whatever service and explain how to reproduce
the error.

Sebastien.

On 09/07/2016 07:10 AM, Sukhraj Singh wrote:
> Please find the Surface Mesh Deformation Plugin Code having the changes
> as attachment.
> I think you should be able to find my changes using some code comparison
> tool. Else let me know.
> It is based on 4.8.1 tag.
>
> A function which is worth noting is as below. In which I get the crash.
> =========================================================================================
>
> //Implement it likewise all mesh vertices as roi
>
>
>    void  set_segment_vertices_as_roi(Polyhedron  mesh)
>
>    {
>
>      std::size_t  number_of_clusters;
>
>      double  smoothing_lambda  ;
>
>
>      //  create  a  property-map  for  segment-ids
>
>      typedef  std::map<Polyhedron::Facet_const_handle,  std::size_t>  
> Facet_int_map;
>
>      Facet_int_map  internal_segment_map;
>
>      boost::associative_property_map<Facet_int_map>  
> segment_property_map(internal_segment_map);
>
>
> //Run the sdf segmentation algo
>
>      int  nb_segments  =  
> SegmentPolyhedron(segment_property_map,number_of_clusters,  smoothing_lambda  
> ,mesh);
>
>
>      std::vector  <Polyhedron>  PolySegments;
>
>      PolySegments.resize(nb_segments);
>
>
> //Get the segement of the mesh
>
>      GetSegments(mesh,segment_property_map,nb_segments,PolySegments,0);
>
>
>
>      vertex_iterator  vb,  ve;
>
>
>
> //Make zeroth segment as ROI
>
>      for(boost::tie(vb, ve) = vertices(PolySegments[0]);  vb  !=  ve;  ++vb)
>
>      {
>
>        //std::cout << __LINE__ << std::endl;
>
>          insert_roi_vertex(*vb); // GIVES CRASH
>
>      }
>
>
>
>
>
>    }
>
> =============================================================================================
>
>
>
> The approach I am following right now is, if you replace the for loop above 
> with the below code. Also commenting the GetSegments() call, as it will not 
> be required.
>
>
>
>
> =============================================================================
>
> for(Polyhedron::Facet_iteratorfacet_it=(*polyhedron()).facets_begin();
> facet_it!=(*polyhedron()).facets_end();++facet_it) {
> if(segment_property_map[facet_it]==segid)// the particular seg id I want
> as ROI
> { //Let'stakethethreeverticesoffacet
> insert_roi_vertex(facet_it->halfedge()->vertex());
> insert_roi_vertex(facet_it->halfedge()->next()->vertex());
> insert_roi_vertex(facet_it->halfedge()->opposite()->vertex());
> }
> }
> ===============================================================================
>
>
>
>
>
>
> Thanks !!
>
>
>
>
> On Wed, Sep 7, 2016 at 2:52 AM, Sebastien Loriot (GeometryFactory)
> <
>  
> <mailto:>>
>  wrote:
>
>     I see no error in this code, which is why a minimal example showing
>     the pb would be great.
>
>     Sebastien.
>
>     On 09/05/2016 03:29 PM, Sukhraj Singh wrote:
>
>         Hi Sebastien,
>            Instead of getting the segment polyhedron and using that to
>         create an
>         ROI, I directly used the segment_map to create the ROI as follows.
>
>         
> =============================================================================
>
>            for(Polyhedron::Facet_iterator  facet_it  =
>         (*polyhedron()).facets_begin();
>
>                           facet_it  !=  (*polyhedron()).facets_end();
>         ++facet_it)
>
>                   {
>
>
>                       if(segment_property_map[facet_it]  ==  segid)//
>         the particular seg id I want as ROI
>
>                       {
>
>                           //Let's  take  the  three  vertices  of  facet
>
>                           insert_roi_vertex(facet_it->halfedge()->vertex());
>
>
>           insert_roi_vertex(facet_it->halfedge()->next()->vertex());
>
>
>           insert_roi_vertex(facet_it->halfedge()->opposite()->vertex());
>
>                       }
>
>
>                   }
>
>         
> ===============================================================================
>
>
>
>         I do get a correct ROI as seen form its vizulization, but have a
>         doubt iterating in this way will push same shared vertex among
>         faces again.
>
>         I have done this because the Segmentation algo implementation
>         returns a facet based property map.
>
>
>         Thanks to clarify.
>
>
>
>         On Mon, Sep 5, 2016 at 4:18 PM, Sebastien Loriot (GeometryFactory)
>         
> <
>  
> <mailto:>
>         
> <mailto:
>  
> <mailto:>>>
>  wrote:
>
>              Could you provide a full example reproducing the issue?
>
>              Thanks,
>
>              Sebastien.
>
>
>              On 09/02/2016 07:24 PM, Sukhraj Singh wrote:
>
>                  Hello,
>
>                  I am trying to create a particular segment (which is
>         the output of
>                  surface mesh segmentation) as ROI in Surface Mesh
>         Deformation.
>
>                  I implemented following code, which is crashing( psb
>         comment below)
>
>                  ==========================================================
>
>                  void  set_all_vertices_as_roi()
>
>                      {
>
>                        vertex_iterator  vb,  ve;
>
>                        for(boost::tie(vb,  ve)  =
>         vertices(*polyhedron());  vb
>                  !=  ve;  ++vb)
>
>                        {
>
>                          insert_roi_vertex(*vb);
>
>                        }
>
>                      }
>
>
>
>                  //Implement it likewise all mesh vertices as roi
>
>
>                      void  set_segment_vertices_as_roi(Polyhedron  mesh)
>
>                      {
>
>                        std::size_t  number_of_clusters;
>
>                        double  smoothing_lambda  ;
>
>
>                        //  create  a  property-map  for  segment-ids
>
>                        typedef  std::map<Polyhedron::Facet_const_handle,
>                  std::size_t>  Facet_int_map;
>
>                        Facet_int_map  internal_segment_map;
>
>                        boost::associative_property_map<Facet_int_map>
>                  segment_property_map(internal_segment_map);
>
>
>                  //Run the sdf segmentation algo
>
>                        int  nb_segments  =
>                  SegmentPolyhedron(segment_property_map,number_of_clusters,
>                  smoothing_lambda  ,mesh);
>
>
>                        std::vector  <Polyhedron>  PolySegments;
>
>                        PolySegments.resize(nb_segments);
>
>
>                  //Get the segement of the mesh
>
>
>
>         GetSegments(mesh,segment_property_map,nb_segments,PolySegments,0);
>
>
>
>                        vertex_iterator  vb,  ve;
>
>
>
>                  //Make zeroth segment as ROI
>
>                        for(boost::tie(vb,  ve)  =
>         vertices(PolySegments[0]);
>                  vb  !=  ve;  ++vb)
>
>                        {
>
>                          //std::cout  <<  __LINE__  <<  std::endl;
>
>                            insert_roi_vertex(*vb); // GIVES CRASH
>
>                        }
>
>
>
>
>
>                      }
>
>
>
>
>         
> ===============================================================================
>
>
>                  Perhaps it is because of size of deform_mesh, which
>         was  equal to
>                  complete non-segmented mesh size. ?
>
>                  There is k-ring selection as well, which is used when
>         picked by
>                  mouse.
>
>                  Can someone suggest me some other way maybe, so that I
>         could
>                  make the
>                  mesh segment as ROI
>
>
>                  Thanks to provide input.
>
>
>                  --
>                  Sukhraj Singh
>
>
>
>              --
>              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>>
>
>
>
>
>
>         --
>         Sukhraj Singh
>
>
>
>     --
>     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>
>
>
>
>
>
> --
> Sukhraj Singh