Interactive rendering of arbitrarily large datasets is a fundamental problem in computer graphics and scientific visualization and a critical capability for many real applications. Interactive visualization of large datasets poses substantial challenges. The visualization pipeline may be broken down into four major stages: retrieval from storage, processing in main memory, rendering in the Graphics Processing Unit (GPU), and display on the screen. The performance of each of these stages is limited by several potential bottlenecks (e.g., disk or network bandwidth, main memory size, GPU triangle throughput, and screen resolution). iRun uses out-of-core data management and speculative visibility prefetching to maintain a working-set of the geometry in memory. Our rendering approach uses GPU-assisted volume rendering with a dynamic set of tetrahedra and uses an out-of-core LOD traversal. Finally, our system is implemented in VTK and allows distributed rendering for high-resolution displays. Using a single commodity PC, our system can render datasets consisting of 14 million tetrahedra while maintaining interactive frame rates.

The generation of a set of point samples is a ubiquitous requirement in many mathematical and computational problems -- from shape statistics, to mesh generation, to visualization. Dynamic particle systems are an intuitive and controllable mechanism for producing very even distributions of points across complex implicit surfaces. Controlled by only a few constraints, these systems can robustly provide nearly-regular packings that smoothly adapt to surface features. The constraints cause particles to first stick to the zero set of an implicit function, and then to move across the surface until particles are arranged in minimal energy configurations. Adaptivity is added into the system by scaling the distance between particles, causing higher densities of particles around surface features. The end result is an adaptive, yet very regular, set of surface points.