Deep Convolutional Reconstruction For Gradient-Domain Rendering

Loading...
Thumbnail Image
Access rights
openAccess
Journal Title
Journal ISSN
Volume Title
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This publication is imported from Aalto University research portal.
View publication in the Research portal
View/Open full text file from the Research portal
Date
2019-07
Major/Subject
Mcode
Degree programme
Language
en
Pages
12
Series
ACM Transactions on Graphics, Volume 38, issue 4
Abstract
It has been shown that rendering in the gradient domain, i.e., estimating finite difference gradients of image intensity using correlated samples, and combining them with direct estimates of pixel intensities by solving a screened Poisson problem, often offers fundamental benefits over merely sampling pixel intensities. The reasons can be traced to the frequency content of the light transport integrand and its interplay with the gradient operator. However, while they often yield state of the art performance among algorithms that are based on Monte Carlo sampling alone, gradient-domain rendering algorithms have, until now, not generally been competitive with techniques that combine Monte Carlo sampling with post-hoc noise removal using sophisticated non-linear filtering. Drawing on the power of modern convolutional neural networks, we propose a novel reconstruction method for gradient-domain rendering. Our technique replaces the screened Poisson solver of previous gradient-domain techniques with a novel densevariant of the U-Net autoencoder, additionally taking auxiliary feature buffers as inputs. We optimize our network to minimize a perceptual image distance metric calibrated to the human visual system. Our results significantly improve the quality obtained from gradient-domain path tracing, allowing it to overtake state-of-the-art comparison techniques that denoise traditional Monte Carlo samplings. In particular, we observe that the correlated gradient samples - that offer information about the smoothness of the integrand unavailable in standard Monte Carlo sampling - notably improve image quality compared to an equally powerful neural model that does not make use of gradient samples.
Description
Keywords
gradient-domain rendering, gradient-domain reconstruction, screened poisson, ray tracing
Other note
Citation
Kettunen , M , Harkonen , E & Lehtinen , J 2019 , ' Deep Convolutional Reconstruction For Gradient-Domain Rendering ' , ACM Transactions on Graphics , vol. 38 , no. 4 , 126 . https://doi.org/10.1145/3306346.3323038