Real-time Monte Carlo Denoising with Weight Sharing Kernel Prediction Network
Authors:
Preprint
Abstract
Real-time Monte Carlo denoising aims at removing severe noise under low samples per pixel (spp) in a strict time budget. Recently, kernel-prediction methods use a neural network to predict each pixel's filtering kernel and have shown a great potential to remove Monte Carlo noise. However, the heavy computation overhead blocks these methods from real-time applications. This paper expands the kernel-prediction method and proposes a novel approach to denoise very low spp (e.g., 1-spp) Monte Carlo path traced images at real-time frame rates. Instead of using the neural network to directly predict the kernel map, i.e., the complete weights of each per-pixel filtering kernel, we predict an encoding of the kernel map, followed by a high-efficiency decoder with unfolding operations for a high-quality reconstruction of the filtering kernels. The kernel map encoding yields a compact single-channel representation of the kernel map, which can significantly reduce the kernel-prediction network's throughput. In addition, we adopt a scalable kernel fusion module to improve denoising quality. The proposed approach preserves kernel prediction methods' denoising quality while roughly halving its denoising time for 1-spp noisy inputs. In addition, compared with the recent neural bilateral grid-based real-time denoiser, our approach benefits from the high parallelism of kernel-based reconstruction and produces better denoising results at equal time.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- A survey on deep learning-based Monte Carlo denoisingComputational Visual Media, 2021
- Multi-resolution terrain rendering using summed-area tablesComputers & Graphics, 2021
- PowerNet: Learning-Based Real-Time Power-Budget RenderingIEEE Transactions on Visualization and Computer Graphics, 2021
- Lightweight Bilateral Convolutional Neural Networks for Interactive Single-Bounce Diffuse Indirect IlluminationIEEE Transactions on Visualization and Computer Graphics, 2020
- A matrix sampling-and-recovery approach for many-lights renderingACM Transactions on Graphics, 2015
- A machine learning approach for filtering Monte Carlo noiseACM Transactions on Graphics, 2015
- Adaptive rendering with non-local means filteringACM Transactions on Graphics, 2012
- On filtering the noise from the random parameters in Monte Carlo renderingACM Transactions on Graphics, 2012
- Amortized supersamplingACM Transactions on Graphics, 2009
- A Review of Image Denoising Algorithms, with a New OneMultiscale Modeling & Simulation, 2005