Animated three dimensional geometry has become a natural part in many areas over the last decade. Industries such as games and special effects have pushed the creative tools and ways processing such content. Animations are typically defined by a series of static meshes where each mesh, or frame, represents a certain time in the animation.
This project work will investigate and implement a method for reconstructing an unstructured mesh sequence with evolving topology. The goal of the method is to increase frame-to-frame coherency of the triangulation. The motivation of the method is that many of current state-of-the-art mesh compression and decimation algorithms for mesh sequences are based on static connectivity.
The method investigated in this thesis is mainly based on the work by presented in Tracking surfaces with evolving topology by Wojtan, Li and Bojsen-Hansen. They use non-rigid alignment to deform a dynamic mesh to track a target mesh sequence. Topological changes such as splitting and merging are handled in two steps. First the mesh
is converted into a signed distance field, discretized on a regular grid. Complex cell tests are used to detect any topological events.
Finally, all the complex cells are cut out of the meshed and re-meshed using the distance field information. To handle robustness issues with cutting algorithm, we present a novel approach for grid based mesh cutting which does not use numerics to decide how to cut the mesh. We conclude that even though the method is appealing in theory and works well for simple cases. The method is, in the current state, not a valid approach
for a production grade system where the input data can be arbitrarily complex.
Source: Linköping University
Author: Birger, Christopher