DataParallel Rasterization of Micropolygons with Defocus and Motion Blur


Kayvon Fatahalian 
Edward Luong  Solomon Boulos  
Kurt Akeley  William R. Mark  Pat Hanrahan  
Presented at High Performance Graphics 2009 
Abstract:
Current GPUs rasterize micropolygons (polygons approximately one pixel in size) inefficiently. We design and analyze the costs of three alternative dataparallel algorithms for rasterizing micropolygon workloads for the realtime domain. First, we demonstrate that efficient micropolygon rasterization requires parallelism across many polygons, not just within a single polygon. Second, we produce a dataparallel implementation of an existing stochastic rasterization algorithm by Pixar, which is able to produce motion blur and depthoffield effects. Third, we provide an algorithm that leverages interleaved sampling for motion blur and camera defocus. This algorithm outperforms Pixar's algorithm when rendering objects undergoing moderate defocus or high motion and has the added benefit of predictable performance.
Paper:
Posted 6/8/2009 