Modeling and visualization of large geometric environments is a popular research area in computer graphics. In this dissertation, a framework for modeling and stereoscopic visualization of large and complex urban environments is presented. The occlusion culling and view-frustum culling is performed to eliminate most geometry that do not contribute to the user's final view. For the occlusion culling process, the shrinking method is employed but performed using a novel Minkowski-difference-based approach. In order to represent partial visibility, a novel data structure called the slice-wise data structure is developed. This data structure is able to represent the preprocessed partial visibility with huge reductions in the storage need. For the stereoscopic visualization, the number of the view-frustum culling operations needed is decreased. The resultant visibility list is rendered using a graphics-processing-unit-based algorithm, which perfectly fits into the proposed slice-wise data structure. The proposed algorithms were implemented on personal computers. Performance experiments show that, the proposed occlusion culling method and the usage of the slice-wise data structure increase the performance by 81 % in frame rates; the graphics-processing-unit-based method increases it by an additional 315 % and decrease storage need by 97 % as compared to occlusion culling using building-level granularity. We show that, a smooth and real-time visualization of large and complex urban environments can be achieved by using the proposed framework.

DATE: 12 June, 2007, Tuesday@ 10:30
PLACE: EA 502