《Ray Casting on Programmable》.pdfVIP

Ray Casting on Programmable Graphics Hardware Martin Kraus Martin Kraus PURPL group, Purdue University PURPL group, Purdue University Overview • Parallel volume rendering with a single GPU • Parallel volume rendering with a single GPU • Implementing ray casting for a GPU • Implementing ray casting for a GPU – Basics – Basics – Optimizations – Optimizations • Published systems • Published systems • Open questions • Open questions 2 Parallel Volume Rendering with a single Graphics Processing Unit (GPU) (1) Even a single GPU is a parallel computer: Even a single GPU is a parallel computer: • Multiple vertex and pixel pipelines • Multiple vertex and pixel pipelines (e.g. “8 pixel pipelines” of QuadroFX and FireGL) (e.g. “8 pixel pipelines” of QuadroFX and FireGL) • APIs support this parallelism • APIs support this parallelism (independent processing of vertices and fragments) (independent processing of vertices and fragments) • When implementing algorithms for a GPU, it’s crucial • When implementing algorithms for a GPU, it’s crucial to exploit this parallelism to exploit this parallelism 3 Parallel Volume Rendering with a single Graphics Processing Unit (GPU) (2) Many volume rendering algorithms cannot Many volume rendering algorithms cannot fully exploit these parallel pixel pipelines: fully exploit these parallel pixel pipelines: • Cell projection: requires visibility sorting of cells • Cell projection: requires visibility sorting of cells • Textured slices: rasterize all fragments of all slices • Textured slices: rasterize all fragments of all slices