Stochastic Pairwise MIS for Unbiased Large‐Kernel Reuse in Real‐Time
Computer Graphics Forum · 14.04.2026
DOI: 10.1111/cgf.70391 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computer Graphics Forum, online erschienen am 14.04.2026. Verfasst von Trevor Hedstrom und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Google, Adobe, Division of Information and Intelligent Systems.
Abstract (Originalsprache)
Spatiotemporal resampling methods such as ReSTIR decrease noise in Monte Carlo rendering of dynamic content by reusing paths across frames and pixels. Standard ReSTIR reuses spatially from a small number of randomly selected neighbors. This reuse suffers when few neighbors contain contributing samples, reducing quality toward that of the underlying path sampler. This commonly occurs during camera or object motion, as regions not present in prior frames are revealed. Increasing the number of spatial neighbors helps but also increases cost. We propose a novel spatial neighbor selection technique, stochastic pairwise MIS, which enables unbiased reuse from many neighbors in real time and focuses reuse on pixels with contributing samples. This provides a significant increase in image quality overall, especially in regions with poor input samples.
Förderung
- Adobe
- Division of Information and Intelligent Systems
Zitieren
APA
Hedstrom, T., Kettunen, M., Lin, D., Wyman, C., & Li, T. (2026). Stochastic Pairwise MIS for Unbiased Large‐Kernel Reuse in Real‐Time. Computer Graphics Forum. https://doi.org/10.1111/cgf.70391
Vancouver
Hedstrom T, Kettunen M, Lin D, Wyman C, Li T. Stochastic Pairwise MIS for Unbiased Large‐Kernel Reuse in Real‐Time. Computer Graphics Forum. 2026. doi:10.1111/cgf.70391
BibTeX
@article{Hedstrom2026,
author = {Hedstrom, Trevor and Kettunen, Markus and Lin, Daqi and Wyman, Chris and Li, Tzu‐Mao},
title = {Stochastic Pairwise MIS for Unbiased Large‐Kernel Reuse in Real‐Time},
journal = {Computer Graphics Forum},
year = {2026},
doi = {10.1111/cgf.70391}
}
EUVIMED