Stochastic Pairwise MIS for Unbiased Large-Kernel Reuse in Real-Time

Simple item page
dc.contributor.authorHedstrom, Trevor
dc.contributor.authorKettunen, Markus
dc.contributor.authorLin, Daqi
dc.contributor.authorWyman, Chris
dc.contributor.authorLi, Tzu-Mao
dc.contributor.editorMasia, Belen
dc.contributor.editorThies, Justus
dc.date.accessioned2026-04-17T14:00:22Z
dc.date.available2026-04-17T14:00:22Z
dc.date.issued2026
dc.description.abstractSpatiotemporal 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.
dc.description.number2
dc.description.sectionheadersLight Transport: Sampling, Waves, and Denoising
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume45
dc.identifier.doi10.1111/cgf.70391
dc.identifier.issn1467-8659
dc.identifier.pages12 pages
dc.identifier.urihttps://diglib.eg.org/handle/10.1111/cgf70391
dc.identifier.urihttps://doi.org/10.1111/cgf.70391
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.
dc.rightsCC-BY-4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectComputing methodologies → Ray tracing
dc.subjectGlobal Illumination
dc.subjectResampling
dc.subjectMultiple Importance Sampling
dc.titleStochastic Pairwise MIS for Unbiased Large-Kernel Reuse in Real-Time
Files
Original bundle
Now showing 1 - 2 of 2
Thumbnail Image
Name:
cgf70391.pdf
Size:
18.33 MB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
paper1324_mm1.zip
Size:
15.27 KB
Format:
Zip file
Download
Collections
45-Issue 2
EG 2026 - Full Papers - CGF 45-Issue 2