A topological comparison of surface extraction algorithms
In many application areas, it is useful to convert the discrete information stored in the nodes of a regular grid into a continuous boundary model. Isosurface extraction algorithms differ on how the discrete information in the grid is generated, on what information does the grid store and on the pro...
| Autores: | , , , , |
|---|---|
| Tipo de recurso: | informe técnico |
| Fecha de publicación: | 2021 |
| País: | España |
| Institución: | Universitat Politècnica de Catalunya (UPC) |
| Repositorio: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglés |
| OAI Identifier: | oai:upcommons.upc.edu:2117/359540 |
| Acceso en línea: | https://hdl.handle.net/2117/359540 |
| Access Level: | acceso abierto |
| Palabra clave: | Three-dimensional imaging Algorithms Surface extraction Marching Cubes algorithm Topologically consistent isosurfaces Rectangular grids Discrete volume models Topology optimization Imatgeria tridimensional Algorismes Àrees temàtiques de la UPC::Informàtica::Infografia |
| Sumario: | In many application areas, it is useful to convert the discrete information stored in the nodes of a regular grid into a continuous boundary model. Isosurface extraction algorithms differ on how the discrete information in the grid is generated, on what information does the grid store and on the properties of the output surface. Recent algorithms offer different solutions for the disambiguation problem and for controlling the final topology. Based on a number of properties of the grid’s grey cells and of the reconstruction algorithms, a characterization of several surface extraction strategies is proposed. The classification presented shows the inherent limitations of the different algorithms concerning global topology control and reconstruction of local features like thin portions of the volume and almost non-manifold regions. These limitations can be observed and are illustrated with some practical examples. We review in light of this classification some of the relevant papers in the literature, and see that they cluster in some areas of the proposed hierarchy, making a case for where it might be more interesting to focus in future research. |
|---|