I'd like to be able to create a TriMesh from non-convex 2D polygons.
I have a PR ready implementing the ear-clipping algorithm. Once we have this, we can think about adding the Hertel-Mehlhorn algorithm that combines the triangles into convex polygons again, to give a good convex decomposition.
The reason for this is that, at least in my experiments, the VHACD algorithm is orders of magnitude slower than this method.
I'd like to be able to create a
TriMeshfrom non-convex 2D polygons.I have a PR ready implementing the ear-clipping algorithm. Once we have this, we can think about adding the Hertel-Mehlhorn algorithm that combines the triangles into convex polygons again, to give a good convex decomposition.
The reason for this is that, at least in my experiments, the VHACD algorithm is orders of magnitude slower than this method.