Harsh Light

Real-time Global Illumination Renderer
based on Voxel Cone Tracing

Jiawen Liang(jiawenl), Yang Zhou(yzhou4)

PROPOSAL

SUMMARY

We are going to implement a real-time global illumination renderer/solution based on hierarchical voxel octree representation of the scene and efficient radiance estimation based on voxel cone tracing.

BACKGROUND

We are going to implement a real-time global illumination renderer/solution based on hierarchical voxel octree representation of the scene and efficient radiance estimation based on voxel cone tracing.

THE CHALLENGE

Data Representation and Query on GPU:  In order to efficiently estimate indirect illumination, a hierarchical voxel tree representation of the scene is construct entirely on GPU which also supports dynamic modification. Each pixel needs to get access to the scene and lighting information efficiently through the structure to perform radiance estimation.

Computation Time:  We would like to achieve interactable frame rate on large scenes with complex geometry. Furthermore, our solution supports dynamic objects and lights.

Divergence:  Although it is generally feasible to parallelize raytracing-like operation, the difference of paths of light often ends up divergence and different memory access patterns.

RESOURCES

We will start the project from scratch.

The deliverable will be running on desktop or laptop (either Windows or Linux) with a moderate CUDA-capable graphics card. The project is implemented using CUDA and OpenGL.

References are listed at the bottom.

GOALS AND DELIVERABLES

The deliverable will be a real-time rendering application that supports limited global illumination on large scenes on desktop. More specifically, it supports:

PLATFORM CHOICE

The project will be implemented in C++ and CUDA, with OpenGL.

The platform will be desktop(either Windows or Linux) with a moderate CUDA-capable graphics card.

SCHEDULE

REFERENCE

Crassin, Cyril, et al. "Interactive indirect illumination using voxel cone tracing." Computer Graphics Forum. Vol. 30. No. 7. Blackwell Publishing Ltd, 2011.

Laine, Samuli, and Tero Karras. "Efficient sparse voxel octrees." IEEE Transactions on Visualization and Computer Graphics 17.8 (2011): 1048-1059.

Crassin, Cyril, et al. "Gigavoxels: Ray-guided streaming for efficient and detailed voxel rendering." Proceedings of the 2009 symposium on Interactive 3D graphics and games. ACM, 2009.