A Diffusion Approach to Radiance Field Relighting using Multi-Illumination Synthesis
Yohan Poirier-Ginter, Alban Gauthier, Julien Philip, Jean-Francois, Lalonde, George Drettakis

TL;DR
This paper presents a novel method for relighting radiance fields from single-illumination multi-view data by leveraging 2D diffusion models to synthesize multi-illumination datasets, enabling realistic 3D scene relighting.
Contribution
It introduces a diffusion-based data augmentation technique and a 3D Gaussian splat representation for relightable radiance fields from limited data.
Findings
Successfully relights synthetic and real scenes under different illumination conditions.
Uses diffusion priors to generate multi-illumination data from single-illumination captures.
Achieves multi-view consistency and realistic relighting results.
Abstract
Relighting radiance fields is severely underconstrained for multi-view data, which is most often captured under a single illumination condition; It is especially hard for full scenes containing multiple objects. We introduce a method to create relightable radiance fields using such single-illumination data by exploiting priors extracted from 2D image diffusion models. We first fine-tune a 2D diffusion model on a multi-illumination dataset conditioned by light direction, allowing us to augment a single-illumination capture into a realistic -- but possibly inconsistent -- multi-illumination dataset from directly defined light directions. We use this augmented data to create a relightable radiance field represented by 3D Gaussian splats. To allow direct control of light direction for low-frequency lighting, we represent appearance with a multi-layer perceptron parameterized on light…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
MethodsDiffusion
