High-Quality Real-Time Rendering Using Subpixel Sampling Reconstruction
This work solves the problem of slow rendering times for high-resolution real-time applications, though it appears incremental as it builds on existing denoising and sampling techniques.
The paper tackles the problem of achieving high-quality real-time rendering by addressing the performance limitations of existing denoising methods, resulting in a method that significantly outperforms previous approaches in denoising quality and reduces time costs, enabling real-time rendering at 2K resolution.
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.