# Real-time Monte Carlo Denoising with Weight Sharing Kernel Prediction Network

arXiv: https://arxiv.org/abs/2202.05977 · HTML: https://mlsanity.com/paper/2202.05977/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Feb 2022.

## Built on by (most cited first, 8 of 8)

- [FuseSR: Super Resolution for Real-time Rendering through Efficient Multi-resolution Fusion](https://mlsanity.com/paper/2310.09726/) (2023)
- [Neural Rendering and Its Hardware Acceleration: A Review](https://mlsanity.com/paper/2402.00028/) (2024)
- [Denoising Monte Carlo Renders with Diffusion Models](https://mlsanity.com/paper/2404.00491/) (2024)
- [Normal and Visibility Estimation of Human Face from a Single Image](https://mlsanity.com/paper/2203.04647/) (2022)
- [An Interactive Image-based Modeling System](https://mlsanity.com/paper/2203.14441/) (2022)
- [High-Quality Real-Time Rendering Using Subpixel Sampling Reconstruction](https://mlsanity.com/paper/2301.01036/) (2023)
- [Gaussian Stippling: Efficient Sorting-Free 3D Gaussian Rendering through Hybrid Sampling and Spatiotemporal Reconstruction](https://mlsanity.com/paper/2609.38488/) (2026)
- [SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering](https://mlsanity.com/paper/2610.05576/) (2026)

## Abstract (arXiv)

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.