Deep Compositional Denoising for High-quality Monte Carlo Rendering

Simple item page
dc.contributor.authorZhang, Xianyaoen_US
dc.contributor.authorManzi, Marcoen_US
dc.contributor.authorVogels, Thijsen_US
dc.contributor.authorDahlberg, Henriken_US
dc.contributor.authorGross, Markusen_US
dc.contributor.authorPapas, Mariosen_US
dc.contributor.editorBousseau, Adrien and McGuire, Morganen_US
dc.date.accessioned2021-07-12T12:08:47Z
dc.date.available2021-07-12T12:08:47Z
dc.date.issued2021
dc.description.abstractWe propose a deep-learning method for automatically decomposing noisy Monte Carlo renderings into components that kernelpredicting denoisers can denoise more effectively. In our model, a neural decomposition module learns to predict noisy components and corresponding feature maps, which are consecutively reconstructed by a denoising module. The components are predicted based on statistics aggregated at the pixel level by the renderer. Denoising these components individually allows the use of per-component kernels that adapt to each component's noisy signal characteristics. Experimentally, we show that the proposed decomposition module consistently improves the denoising quality of current state-of-the-art kernel-predicting denoisers on large-scale academic and production datasets.en_US
dc.description.number4
dc.description.sectionheadersDenoising
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume40
dc.identifier.doi10.1111/cgf.14337
dc.identifier.issn1467-8659
dc.identifier.pages1-13
dc.identifier.urihttps://doi.org/10.1111/cgf.14337
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf14337
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectComputing methodologies --> Ray tracing
dc.subjectNeural networks
dc.titleDeep Compositional Denoising for High-quality Monte Carlo Renderingen_US
Files
Original bundle
Now showing 1 - 5 of 10
Thumbnail Image
Name:
v40i4pp001-013.pdf
Size:
56.55 MB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
index.html
Size:
25.46 KB
Format:
Hypertext Markup Language
Download
Thumbnail Image
Name:
paper1010_mm.pdf
Size:
264.65 KB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
paper1010_mm_crc4.zip
Size:
326.41 MB
Format:
Zip file
Download
No Thumbnail Available
Name:
paper1010_mm_crc5.zip
Size:
119.91 MB
Format:
Zip file
Download
Collections
40-Issue 4