CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

On 09/13/2017 01:32 PM, Iasonm wrote:
> No I did not, but the thing is that a skeleton is not a 2-manifold surface
> right? So maybe that's why edge_collapse was not used  here
> <https://github.com/CGAL/cgal/blob/master/Surface_mesh_skeletonization/include/CGAL/internal/Surface_mesh_skeletonization/Curve_skeleton.h>
> . Maybe I should use a similar approach. Is there any doc regarding
> Curve_skeleton.h or any document that describes its implementation (besides
> the source code)?

If you're talking about the final step when the meso-skeleton is turned 
into a curve-skeleton, then indeed collapse_edge is not used and
the FaceGraph model is no longer used (another data structure is used
for collapsing those edges). I think this step is explained in a former
paper:
A U O. K.-C., T AI C.-L., C HU H.-K., C OHEN -O R
D., L EE T.-Y.: Skeleton extraction by mesh contraction. ACM
Transactions on Graphics (Proc. of SIGGRAPH) 27, 3 (2008)

HTH

Sebastien.


> Best Regards,
> Iason
>
>
>
> --
> Sent from: http://cgal-discuss.949826.n4.nabble.com/