Re: how to group a set of triangles based on its original geometry model

You are right that it would be easier if I keep the parenting
information from the meshing process. Unfortunately the mesh is
generated by using a third party software. There is no parenting
information attach to it.

Therefore I need to regroup them properly myself.

jeff

On Jul 23, 3:31 pm, Hans-Bernhard Bröker <HBBroe...@xxxxxxxxxxx>
wrote:
jeff wrote:
I am trying to identify and group a set of triangle based its
corresponding geometry model.

I think you're trying to tackle this from the wrong end. It would be
far easier not to throw away that information during triangle
decomposition of the original model data.

Let's say I have 3 models, a cylinder, a cone and a cube. All these 3
models undergone surface meshing process and generate a common list of
the triangles.

So why did you throw away the information which triangle came from which
primitive?

I need to isolate the triangles into a number of group(in this case it
would be 3).

That's generally impossible. How is an algorithm supposed to know
whether you wanted a cylinder or a wall plus two polygons that just
happen to share two polygonal edges?


.