Two-Stage Monte Carlo Denoising with Adaptive Sampling and Kernel Pool

TL;DR

This paper introduces a two-stage Monte Carlo denoising method that adaptively adjusts sampling and uses a kernel pool for improved spatial-temporal stability, outperforming existing techniques.

Contribution

The proposed approach combines adaptive sampling with a kernel pool and novel pooling operators to enhance denoising quality in Monte Carlo rendering.

Findings

Outperforms state-of-the-art denoising methods in numerical error

Achieves better visual quality in rendered images

Validated on both Mitsuba and custom RTX-based scenes

Abstract

Monte Carlo path tracer renders noisy image sequences at low sampling counts. Although great progress has been made on denoising such sequences, existing methods still suffer from spatial and temporary artifacts. In this paper, we tackle the problems in Monte Carlo rendering by proposing a two-stage denoiser based on the adaptive sampling strategy. In the first stage, concurrent to adjusting samples per pixel (spp) on-the-fly, we reuse the computations to generate extra denoising kernels applying on the adaptively rendered image. Rather than a direct prediction of pixel-wise kernels, we save the overhead complexity by interpolating such kernels from a public kernel pool, which can be dynamically updated to fit input signals. In the second stage, we design the position-aware pooling and semantic alignment operators to improve spatial-temporal stability. Our method was first benchmarked on 10 synthesized scenes rendered from the Mitsuba renderer and then validated on 3 additional scenes rendered from our self-built RTX-based renderer. Our method outperforms state-of-the-art counterparts in terms of both numerical error and visual quality.