We present a general framework for post-processing and optimizing surface meshes with respect to various target criteria. On the one hand, the framework allows us to control the shapes of the mesh triangles by applying simple averaging operations; on the other hand we can control the Hausdorff distance to some reference geometry by minimizing a quadratic energy. Due to the simplicity of this setup, the framework is efficient and easy to implement, yet it also constitutes an effective and versatile tool with a variety of possible applications. In particular, we use it to reduce the texture distortion in animated mesh sequences, to improve the results of cross-parameterizations, and to minimize the distance between meshes and their remeshes.
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
- Computer Graphics and Computer-Aided Design
- Hausdorff distance
- Laplacian smoothing
- Mesh optimization