Presentation / Installation


Neural Partitioning Pyramids for Denoising Monte Carlo Renderings
SessionReal-Time Rendering: Gotta Go Fast!
DescriptionWe present a novel low-pass pyramidal kernel prediction scheme to denoise path-traced images. We learn partitioning the input radiance between pyramidal layers and apply learnable denoising and upsampling kernels. Our scheme is energy-preserving and numerically stable, resolving artefacts of previous filters. Our method scales from interactive to offline applications.
Authors
Martin Balint
Max Planck Institute for Informatics
Krzysztof Wolski
Max Planck Institute for Informatics
Karol Myszkowski
Max Planck Institute for Informatics
Hans-Peter Seidel
Max Planck Institute for Informatics
Rafał Mantiuk
University of Cambridge
Event Type
Technical Paper
TimeThursday, 10 August 20239:43am - 9:53am PDT
LocationRoom 515 A
ACM Digital Library Technical Paper PDF
Session Time & Location
Sunday, 6 August 20236pm - 8:30pm PDTWest Hall B
Thursday, 10 August 20239am - 10:30am PDTRoom 515 A
Interest Areas
Research & Education
Recordings
Livestreamed
Recorded
Keywords
Image Processing
Real-Time
Rendering
Video
Registration Categories
FC
FCS
V
VS
EFC
Next PresentationNext Presentation
Deep Real-time Volumetric Rendering Using Multi-feature Fusion