Time and Place:
Preliminary session (Vorbesprechung) on Monday, Oct 22, 2007 (14:00) in room 02.05.033. The seminar session will be two block sessions on Dec 6 and Dec 8 (see schedule below).
A quite common approach to represent continuous data in 2D, 3D, or even higher-dimensional domains is to approximate or interpolate this data on an appropriately generated computational grid. Applications exist in many fields of computational sciences; typical examples are finite difference or finite element methods in numerical simulation, the representation of geometrical objects in computer graphics or visualisation, and even general classification problems in data mining.
Algorithms and methods for grid generation are at least as diverse as their applications. It is the aim of this seminar to discuss several common approaches, for example:
The talks (approx. 45 minutes) in the seminar will be given in English. In addition to her or his talk, each student will have to prepare a short paper (8-10 pages), which will be reviewed by two other seminar participants. The paper and the two reviews will, in addition to the talk, be considered to determine the final grade.
|Thursday, Dec 6, 2007 (14:30)
|Hierarchical basis representation and approximation of functions on sparse grids||Bernhard Gatzhammer|
|Classification and data mining on sparse grids||Janos Benk|
|Sparse grids and the combination technique||Deepak Pandey|
|Tuesday, Dec 11, 2007 (15:00)
|Octree generation for simple objects and CSG||Asim Önder|
|Spacetree meshes and space-filling Curves||Michael Lieb|
|Friday, Dec 14, 2007 (14:00)
Triangular and Tetrahedral Grids
|Delauney triangulation and advancing front methods||Sumit Paranjape|
|Data structures and algorithms for adaptive triangular grids||Bian Xin|
|Tetrahedral (3D) meshes based on bisection||Piyush Gupta|