Lecture 8: Mesh Representations and Geometry Processing (67)
aliwang2001
I was wondering why we would use quad meshes over triangle meshes, and vice versa. Here's some stuff I found:
Quad meshes are a lot easier to work with and are a lot more efficient for subdivision operations, producing smoother results with fewer artifacts. However, the generation of quad meshes is difficult and requires more computational power, so triangle meshes is best for rendering.
kkkhanl
I was originally wondering the same thing as @aliwang. Now I'm wondering if there is a simple way to convert from one to another (first generate triangle meshes, then convert to quad meshes?)
justin-shao
@kkkhanl Conversion from quad meshes to triangle meshes is trivial since you can subdivide each quad into two triangles. However, the reverse is sometimes impossible since the number of faces (triangles) is not guaranteed to be even. This probably means that some triangles can remain by then end of the conversion.
I was wondering why we would use quad meshes over triangle meshes, and vice versa. Here's some stuff I found: Quad meshes are a lot easier to work with and are a lot more efficient for subdivision operations, producing smoother results with fewer artifacts. However, the generation of quad meshes is difficult and requires more computational power, so triangle meshes is best for rendering.
I was originally wondering the same thing as @aliwang. Now I'm wondering if there is a simple way to convert from one to another (first generate triangle meshes, then convert to quad meshes?)
@kkkhanl Conversion from quad meshes to triangle meshes is trivial since you can subdivide each quad into two triangles. However, the reverse is sometimes impossible since the number of faces (triangles) is not guaranteed to be even. This probably means that some triangles can remain by then end of the conversion.