Perceptual error optimization for Monte Carlo animation rendering

ACM SIGGRAPH Asia 2023 (Conference Track)

We propose an optimization framework to obtain perceptually pleasing error distribution in Monte Carlo animation rendering. The output of our algorithm is a sample set spanning multiple image pixels and frames. Here we show an image of a 30-frame sequence rendered with 1 sample/pixel per frame. We display a version of the animation filtered temporally using the kernel of Mantiuk et al. [2021], to mimic its perception at one time instant. On the right we a show spatial (XY) crop and a spatio-temporal (XT) slice, along with the power spectra (DFT) of their corresponding error images. Our error distribution exhibits better blue-noise properties than that of previous work [Wolfe et al. 2022], also reflected in the perceptual error metric reported on the left.


Independently estimating pixel values in Monte Carlo rendering results in a perceptually sub-optimal white-noise distribution of error in image space. Recent works have shown that perceptual fidelity can be improved significantly by distributing pixel error as blue noise instead. Most such works have focused on static images, ignoring the temporal perceptual effects of animation display. We extend prior formulations to simultaneously consider the spatial and temporal domains, and perform an analysis to motivate a perceptually better spatio-temporal error distribution. We then propose a practical error optimization algorithm for spatio-temporal rendering and demonstrate its effectiveness in various configurations.

Fast-Forward Video

Presentation Video

Note that video takes time to download as it is fairly large. This is due to less aggressive video compression encoding, which had to be used in order to keep noise visible.

BibTeX reference

author = {Mi\v{s}a Kora\'{c} and Corentin Sala\"{u}n and Iliyan Georgiev and Pascal Grittmann and Philipp Slusallek and Karol Myszkowski and Gurprit Singh},
title = {Perceptual error optimization for Monte Carlo animation rendering},
booktitle = {ACM SIGGRAPH Asia 2023 Conference Proceedings},
year = {2023},
doi = {10.1145/3610548.3618146},
isbn = {979-8-4007-0315-7/23/12}


This project has received funding from the European Union’s Horizon 2020 research and innovation program under Marie SkłodowskaCurie grant agreement no. 956585. We thank the anonymous reviewers for their feedback, and the authors of the following scenes: julioras3d (Chopper), NewSee2l035 (Modern Hall), Benedikt Bitterli (Utah Teapot), Wig42 (Living Room), and Jay Hardy (White Room).