Baking Relightable NeRF for Real-time Direct/Indirect Illumination Rendering

TL;DR

This paper introduces a real-time relighting method for NeRF-based rendering that efficiently handles direct and indirect illumination using CNN and hash grid-based renderers, enabling photo-realistic view synthesis under new lighting conditions.

Contribution

It presents a novel combination of CNN and hash grid-based renderers trained via distillation for fast, accurate relighting of NeRF models with minimal quality loss.

Findings

Achieves real-time relighting with negligible quality loss.

Effectively handles both direct and indirect illumination.

Enables photo-realistic view synthesis under unseen lighting.

Abstract

Relighting, which synthesizes a novel view under a given lighting condition (unseen in training time), is a must feature for immersive photo-realistic experience. However, real-time relighting is challenging due to high computation cost of the rendering equation which requires shape and material decomposition and visibility test to model shadow. Additionally, for indirect illumination, additional computation of rendering equation on each secondary surface point (where reflection occurs) is required rendering real-time relighting challenging. We propose a novel method that executes a CNN renderer to compute primary surface points and rendering parameters, required for direct illumination. We also present a lightweight hash grid-based renderer, for indirect illumination, which is recursively executed to perform the secondary ray tracing process. Both renderers are trained in a distillation from a pre-trained teacher model and provide real-time physically-based rendering under unseen lighting condition at a negligible loss of rendering quality.