Fast CPU Volume Rendering with Coherent BVH Traversal
Participants: Aaron Knoll, Computational Postdoc Fellow, Futureslab, Mathematics and Computer Science at Argonne National Laboratory
Room 2068, Engineering Research Facility
Guest speaker Aaron Knoll presents his research of “Fast CPU Volume Rendering with Coherent BVH Traversal” at EVL’s tech meeting, Wed 1/26/11, 11am.
We present an efficient method for volume rendering by raycasting on the CPU. We employ coherent packet traversal of an implicit bounding volume hierarchy, heuristically pruned using preintegrated transfer functions, to exploit empty or homogeneous space. We also detail SIMD optimizations for volumetric integration, trilinear interpolation, and gradient lighting. The resulting system performs well on low-end and laptop hardware, and can outperform out-of-core GPU methods by orders of magnitude when rendering large volumes without level-of-detail (LOD) on a workstation. We show that, while slower than GPU methods for low-resolution volumes, an optimized CPU renderer does not require LOD to achieve interactive performance on large data sets.
Aaron Knoll is a Computational Postdoc Fellow with the Futureslab visualization group in Mathematics and Computer Science at Argonne National Laboratory. He received his MS in Electrical Engineering from the University of Essex in the UK, and his PhD in Computer Science in 2009 from the SCI Institute at the University of Utah. Previously, he worked in game development, and in the graphics research group at Intel Corporate Technology Group (CTG). His research interests include volume visualization, interactive ray tracing, and combining visualization and analysis in large-scale scientific computing.
Date: January 26, 2011