RL-based Stateful Neural Adaptive Sampling and Denoising for Real-Time Path Tracing

TL;DR

This paper introduces a reinforcement learning-based framework for neural adaptive sampling and denoising in real-time path tracing, significantly improving visual quality and reducing rendering times.

Contribution

It presents a novel end-to-end trainable system combining importance sampling, latent space encoding, and denoising, optimized via reinforcement learning for real-time rendering.

Findings

Achieves 1.6x faster rendering at equal quality.

Improves visual quality on challenging datasets.

Reduces noise effectively in real-time path tracing.

Abstract

Monte-Carlo path tracing is a powerful technique for realistic image synthesis but suffers from high levels of noise at low sample counts, limiting its use in real-time applications. To address this, we propose a framework with end-to-end training of a sampling importance network, a latent space encoder network, and a denoiser network. Our approach uses reinforcement learning to optimize the sampling importance network, thus avoiding explicit numerically approximated gradients. Our method does not aggregate the sampled values per pixel by averaging but keeps all sampled values which are then fed into the latent space encoder. The encoder replaces handcrafted spatiotemporal heuristics by learned representations in a latent space. Finally, a neural denoiser is trained to refine the output image. Our approach increases visual quality on several challenging datasets and reduces rendering times for equal quality by a factor of 1.6x compared to the previous state-of-the-art, making it a promising solution for real-time applications.