You are viewing the course site for a past offering of this course. The current offering may be found here.
Lecture 8: Meshes and Geometry Processing (12)
zsano1

IN this case, we only store the vertices of triangles, so a single line cannot be represented.

fywu85

Furthermore, there will be a lot of book keeping that one should do if they want to transform, reduce, or increase the mesh. For instance, to apply an affine transformation to the mesh, one needs to transform all the points in the data structure, two thirds of which are redundant.

qqqube

How would edge-splitting and edge-flipping operations work in this case? I'm assuming you'd have to traverse the entire list, compare all the edges until you find mutual ones, and then update the respective triangles (which seems extremely inefficient).

avinashnandakumar

@qqqcube I think so, if you were given a specific triangle and two vertices to flip or split, we could directly go to it because of the hashmap but then we would have to find the mutual vertices that are shared by triangles since we couldn't depend on knowing which triangles are neighbor triangles. And then we would have to keep references to all these pointers and reassign. Seems way too convoluted!

You must be enrolled in the course to comment