Publications

Interactive Path Tracing and Reconstruction of Sparse Volumes

We combine state-of-the-art techniques into a system for high-quality, interactive rendering of participating media. We leverage unbiased volume path tracing with multiple scattering, temporally stable neural denoising and NanoVDB, a fast, sparse voxel tree data structure for the GPU, to explore what performance and image quality can be obtained for rendering volumetric data. Additionally, we integrate neural adaptive sampling to significantly improve image quality at a fixed sample budget. Our system runs at interactive rates at 1920 x 1080 on a single GPU and produces high quality results for complex dynamic volumes.

 

Authors
Nikolai Hofmann (NVIDIA)
Publication Date
Uploaded Files
Preprint7.01 MB
Video250.34 MB
Award
Best Paper Presentation, i3D 2021