High-Quality Real-Time Rendering Using Subpixel Sampling Reconstruction

TL;DR

This paper introduces a novel Monte Carlo sampling strategy and a subpixel sampling reconstruction denoiser that together enable real-time high-quality rendering at 2K resolution, outperforming previous methods in speed and quality.

Contribution

The paper presents a new Monte Carlo sampling method and a dedicated denoiser that together significantly improve real-time rendering quality and speed at high resolutions.

Findings

Outperforms previous denoising methods in quality

Reduces overall rendering time significantly

Enables real-time 2K resolution rendering

Abstract

Generating high-quality, realistic rendering images for real-time applications generally requires tracing a few samples-per-pixel (spp) and using deep learning-based approaches to denoise the resulting low-spp images. Existing denoising methods have yet to achieve real-time performance at high resolutions due to the physically-based sampling and network inference time costs. In this paper, we propose a novel Monte Carlo sampling strategy to accelerate the sampling process and a corresponding denoiser, subpixel sampling reconstruction (SSR), to obtain high-quality images. Extensive experiments demonstrate that our method significantly outperforms previous approaches in denoising quality and reduces overall time costs, enabling real-time rendering capabilities at 2K resolution.