GeometryProcessing/Spring2009/Schedule/Surface Representations

• John Meier

I thought Thursday's lecture and the assigned reading provided a good definition of the meaning of "surface representation" in the context of the course. Since my studies to this point have dealt mostly with practicing rendering techniques, I haven't had much cause to think about the effectiveness of the representation of the geometric data (especially for purposes of modification/smoothing/compression/storage). So I feel like the surface representation material is helping me get my thinking oriented in the right direction for this course, or considering the actual structure and properties of the geometric data rather than just how to visualize it. I have run across many of the topics from the lecture in varying degrees of detail (parametric/implicit representations, splines, octrees/BSPs, compression as distance from predicted value, etc.) and it was good to get a brief review of their definitions, but I will definitely need to pull out some textbooks from earlier courses if we're going to be coding them from scratch ourselves (i.e., spline surfaces or mesh compression schemes). I'm interested in the half-edge mesh data structure and seeing how it can be used (I assume from the further reading that we might be using the OpenMesh library?), and while I have an intuitive grasp of the Edgebreaker compression algorithm, I'm still a bit unsure on the details of its operation.