The Community for Technology Leaders
Green Image
A proposed method reduces noise in stochastic ray tracing for interactive progressive rendering. The method accumulates high-variance light paths in a separate buffer, which is filtered by a high-quality edge-preserving filter. Then, this method adds a combination of the noisy unfiltered samples and the less noisy (but biased) filtered samples to the low-variance samples to form the final image. A novel per-pixel blending operator combines both contributions in a way that respects a user-defined threshold on perceived noise. This method can provide fast, reliable previews, even in the presence of complex features such as specular surfaces and high-frequency textures. At the same time, it's consistent in that the bias due to filtering vanishes in the limit.
Noise abatement, Rendering (computer graphics), Noise level, Interactive systems, Image edge detection, Lighting, Noise measurement, Ray tracing

K. Schwenk, A. Kuijper, J. Behr and D. W. Fellner, "Practical Noise Reduction for Progressive Stochastic Ray Tracing with Perceptual Control," in IEEE Computer Graphics and Applications, vol. 32, no. 6, pp. 46-55, 2012.
247 ms
(Ver 3.3 (11022016))