We present a lightweight neural representation to encode the spatially-varying distributions, serving a novel alternative for path guiding algorithms. Our method could be efficiently trained with gradient-based optimization strategies and is practical for parallelization on the GPUs.
Abstract
Previous path guiding techniques typically rely on spatial subdivision structures to approximate directional target distributions, which may cause failure to capture spatio-directional correlations and introduce parallax issue. In this paper, we present Neural Parametric Mixtures (NPM), a neural formulation to encode target distributions for path guiding algorithms. We propose to use a continuous and compact neural implicit representation for encoding parametric models while decoding them via lightweight neural networks. We then derive a gradient-based optimization strategy to directly train the parameters of NPM with noisy Monte Carlo radiance estimates. Our approach efficiently models the target distribution (incident radiance or the product integrand) for path guiding, and outperforms classical guiding methods based on explicit trees by capturing the spatio-directional correlations more accurately. Moreover, our approach is more training efficient and is practical for parallelization on modern GPUs.
We also refer readers to a concurrent work Online Neural Path Guiding with Normalized Anisotropic Spherical Gaussians which contains a similar idea and comprehensive experiments.