Neural Temporal Adaptive Sampling and Denoising

Simple item page
dc.contributor.authorHasselgren, Jonen_US
dc.contributor.authorMunkberg, Jacoben_US
dc.contributor.authorSalvi, Marcoen_US
dc.contributor.authorPatney, Anjulen_US
dc.contributor.authorLefohn, Aaronen_US
dc.contributor.editorPanozzo, Daniele and Assarsson, Ulfen_US
dc.date.accessioned2020-05-24T12:51:20Z
dc.date.available2020-05-24T12:51:20Z
dc.date.issued2020
dc.description.abstractDespite recent advances in Monte Carlo path tracing at interactive rates, denoised image sequences generated with few samples per-pixel often yield temporally unstable results and loss of high-frequency details. We present a novel adaptive rendering method that increases temporal stability and image fidelity of low sample count path tracing by distributing samples via spatio-temporal joint optimization of sampling and denoising. Adding temporal optimization to the sample predictor enables it to learn spatio-temporal sampling strategies such as placing more samples in disoccluded regions, tracking specular highlights, etc; adding temporal feedback to the denoiser boosts the effective input sample count and increases temporal stability. The temporal approach also allows us to remove the initial uniform sampling step typically present in adaptive sampling algorithms. The sample predictor and denoiser are deep neural networks that we co-train end-to-end over multiple consecutive frames. Our approach is scalable, allowing trade-off between quality and performance, and runs at near real-time rates while achieving significantly better image quality and temporal stability than previous methods.en_US
dc.description.number2
dc.description.sectionheadersDeep Learning for Rendering
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume39
dc.identifier.doi10.1111/cgf.13919
dc.identifier.issn1467-8659
dc.identifier.pages147-155
dc.identifier.urihttps://doi.org/10.1111/cgf.13919
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf13919
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectComputing methodologies
dc.subjectNeural networks
dc.subjectRay tracing
dc.titleNeural Temporal Adaptive Sampling and Denoisingen_US
Files
Original bundle
Now showing 1 - 5 of 5
Thumbnail Image
Name:
v39i2pp147-155.pdf
Size:
17.56 MB
Format:
Adobe Portable Document Format
Download
Thumbnail Image
Name:
supplemental.pdf
Size:
12.12 MB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
image_comparisons.zip
Size:
154.25 MB
Format:
Zip file
Download
No Thumbnail Available
Name:
readme.txt
Size:
363 B
Format:
Plain Text
Download
No Thumbnail Available
Name:
video.mp4
Size:
252.64 MB
Format:
Unknown data format
Download
Collections
39-Issue 2