high performance computing on graphics processing units: hgpu.org

Volume rendering has always been a popular topic in computer graphics as it captures more realistic rendering results. However, most of the rendering targets are focused on participating media distributed in the scene. In this project we integrate an Octree data structure with our path tracer in CUDA to achieve efficient volume rendering of models. Our efforts are put on the implementation of a compact model representation in Octree with an efficient parametric algorithm for tree traversal. With the proposed Octree combined into our CUDA path tracer, the volume rendering by each ray per thread is made possible for average Nvidia graphics cards. Our rendering algorithm extends the classic ray tracing process into path tracing with light transport in participating media, which gathers light energy in a physically based manner. Finally, the global illumination examples and comparisons of cloud like models are shown.