IllumiNeRF: 3D Relighting Without Inverse Rendering
This addresses the challenge of computationally expensive and brittle inverse rendering for relighting in computer vision, offering a simpler and competitive alternative.
The paper tackles the problem of relightable 3D view synthesis from images under unknown lighting by proposing a method that avoids inverse rendering, instead using an image diffusion model to relight input images and then reconstructing a Neural Radiance Field, achieving state-of-the-art results on multiple benchmarks.
Existing methods for relightable view synthesis -- using a set of images of an object under unknown lighting to recover a 3D representation that can be rendered from novel viewpoints under a target illumination -- are based on inverse rendering, and attempt to disentangle the object geometry, materials, and lighting that explain the input images. Furthermore, this typically involves optimization through differentiable Monte Carlo rendering, which is brittle and computationally-expensive. In this work, we propose a simpler approach: we first relight each input image using an image diffusion model conditioned on target environment lighting and estimated object geometry. We then reconstruct a Neural Radiance Field (NeRF) with these relit images, from which we render novel views under the target lighting. We demonstrate that this strategy is surprisingly competitive and achieves state-of-the-art results on multiple relighting benchmarks. Please see our project page at https://illuminerf.github.io/.