In this report, an object-space solution to occluder fusion of OBB occluders is explored. Two different approaches are considered were the object-space fusion is reduced to a 2D problem. The first approach finds axis-aligned silhouettes within the projection of occluder OBBs which are then fused together creating large axis-aligned silhouettes. The other approach creates concave hulls of the projected OBB silhouettes from which convex inscribed silhouettes are then found. These silhouettes are then converted back to object-space where shadow frusta created around the silhouettes are used for the culling operation. The effectiveness of the two approaches is evaluated considering the amount of culled geometry. It is shown that fused convex silhouettes are needed to produce competitive results.
This paper presents techniques for speeding up commonly used algorithms forbounding volume (BV) computation, such as the AABB, sphere and k-DOP. Byexploiting the possibilities of parallelismin modern processors, the result exceedsthe expected theoretical result. The methods focus on data-level-parallelism(DLP) using Intel’s SSE instructions, for operations on 4 parallel independentsingle precision floating point values, with a theoretical speed-up factor of 4 ondata throughput. Still, a speed-up between 7–9 are shown in the computation ofAABBs and k-DOPs. For the computation of tight fitting spheres the speed-upfactor halts at approximately 4 due to a limiting data dependency. In addition,further parallelization by multithreading algorithms on multi-core CPUs showsspeed-up factors of 14 on 2 cores and reaching 25 on 4 cores, compared to nonparallel algorithms.