I’m trying to shadow a complex geometry using shadow volume technique.

The model is a non-closed mesh. It’s made up of several separate objects composed of triangulated planar surfaces.

Therefore instead of computing the silhuette of the entire object, I decided to compute the silhouette for each triangle facing the light, and construct a shadow volume of this triangle.

Hence I end up with many triangular intersecting shadow volumes.

My question is, does this work with no artifacts, since the shadow volume technique handles multiple shadow volumes, and what I have now is pretty much multiple shadow volumes but based on one (occluder) using it’s triangles?

Thanks.