GigaWalk: Interactive Walkthrough of Complex Environments

We present a new parallel algorithm for interactive walkthrough of complex, gigabyte-sized environments. Our approach combines occlusion culling and levelsof-detail and uses two graphics pipelines with one or more processors. We use a unified scene graph representation for multiple acceleration techniques, and we present novel algorithms for clustering geometry spatially, computing a scene graph hierarchy, performing conservative occlusion culling, and performing loadbalancing between graphics pipelines and processors. The resulting system, GigaWalk, has been used to render CAD environments composed of tens of millions of polygons at interactive rates on an SGI Onyx system with two Infinite Reality rendering pipelines. Overall, our system’s combination of levels-of-detail and occlusion culling techniques results in significant improvements in frame-rate over view-frustum culling or either single technique alone.