Progressive Spatiotemporal Variance-Guided Filtering

Jan Dundr

Supervisor(s): doc. Ing. Jiří Bittner, Ph.D.

Czech Technical University


Abstract: Path tracing is still very hard to do in real-time due to its complex recursive light interactions: even the newest GPUs struggle to keep real-time framerates for more than just a few samples per pixel, which can lead to a very noisy output or even no useful data for some more problematic areas. This paper uses a recent approach to processing the resulting image: demodulated samples are accumulated from previous frames using reprojection and subsequently filtered by a fast bilateral filter constrained by normals, depth, and variance (which will stop blurring valuable details). This results in a temporally stable noise-free output which converges in a few frames. We implemented the method using OpenGL and incorporated it in an existing high-performance CPU path tracer. We extended the method by putting it in the progressive rendering framework, where initially less than one sample per pixel is shot to increase interactivity. We evaluate the performance and visual quality of this algorithm in several test cases with mostly diffuse illumination.
Keywords: Graphics Hardware, Image Processing, Real-time Graphics, Rendering, Video Games
Full text:
Year: 2018