filename     : Zhan23a.pdf
entry        : inproceedings
conference   : Eurographics Symposium on Rendering, Industry Track
pages        : 
year         : 2023
month        : June
title        : Deep Compositional Denoising on Frame Sequences
subtitle     :
author       : Xianyao Zhang, Gerhard Roethlin, Marco Manzi, Markus Gross, Marios Papas
booktitle    : 
ISSN/ISBN    : 
editor       : Tobias Ritschel and Andrea Weidlich 
publisher    : The Eurographics Association
publ.place   : Eurographics Digital Library
volume       : 
issue        : 
language     : English
keywords     : Denoising, Ray tracing, Neural Networks
abstract     : We propose a novel extension for processing frame sequences that provides context from neighboring frames while avoiding redundant computation and a novel quantization approach that reduces the memory overhead without introducing artifacts. Our method achieves a significant improvement in denoising quality over KPAL and ameliorates temporal denoising artifacts from the single-frame approach. This enables the use of the compositional denoiser in production settings.