Date of Award
2015
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Computer Science
Committee Chair
Chao Peng
Committee Member
Timothy S. Newman
Committee Member
Vinny Argentina
Subject(s)
Finite element method, Morphing (Computer animation), Numerical grid generation (Numerical analysis)
Abstract
This study presents an approach for establishing surface correspondence between multiple genus zero triangle meshes and animating a morph between them. The concepts of mesh simplification and refinement are used to progressively parameterize meshes using a novel metric for measurement of distortion in parameterized triangles. The parameterization algorithm is quite efficient and can process tens of thousands of triangles per minute. Feature correspondence is established between meshes through a two-step feature alignment procedure that aligns user selected features in the parametric domain. The parameterizations are used to merge the shapes from input models in order to generate a single mesh that can be deformed to look like any of the input models. The remeshing algorithm ensures adequate vertex density in appropriate regions in order to preserve features from all input models while also maintaining a minimal number of components (vertices/triangles) so that morphing in real-time is feasible. The presented approach is also able to blend the textures applied to the original models during morphing. A novel technique is presented for re-organizing texture charts and reassigning texture coordinates in order to avoid having to interpolate texture coordinates across texture borders during remeshing. The techniques developed in this study can handle complex meshes, and can produce visually pleasing results with non-trivial feature alignment. Comparisons with other related works in terms of efficiency and quality are also presented in this thesis. The final section exposes limitations in the proposed techniques, and suggests ways to overcome some of them.
Recommended Citation
Timalsena, Sabin, "Efficient spherical parameterization for feature aligned mesh morphing" (2015). Theses. 126.
https://louis.uah.edu/uah-theses/126