Procedure Mesh Repairing using Deep Networks
BT/MT
Status | open |
Student | N/A |
Advisor | Changkun Ou |
Professor | Prof. Dr. Andreas Butz |
Task
Description
Geometry processing is an emerging field for processing 3D domain data, such as polygon mesh, point clouds, voxels. In practice, the processed model are either used by downstream algorithm processing or artists for manually editing.
Unfortuanately, neither algorithms or 3D artist guarantees to produce polygon meshes without flaws. Due to the recent developments of deep neural networks, it is interesting to see how these advances can be applied to resolve artifacts repairing.
Tasks
- Perform a literature review
- Collect polygon meshes that contains flaws and artifacts
- Explore and design methods to deal with faulty mesh inputs
- Summarize your findings in a thesis and presenting them to audiences
Requirements
- Good communication skills in English
- Have experience or projects using Blender and Python
- General knowledge about machine learning and computer graphics