LumiNet: Latent Intrinsics Meets Diffusion Models for Indoor Scene Relighting

TL;DR

LumiNet introduces a novel architecture combining generative models and latent intrinsic representations to transfer complex lighting effects in indoor scene relighting, outperforming existing methods.

Contribution

It proposes a new training data curation strategy and a modified diffusion-based ControlNet that processes dual latent representations for effective lighting transfer.

Findings

Successfully transfers complex lighting phenomena including specular highlights.

Outperforms existing approaches on challenging indoor scenes.

Preserves scene geometry and albedo while transferring lighting.

Abstract

We introduce LumiNet, a novel architecture that leverages generative models and latent intrinsic representations for effective lighting transfer. Given a source image and a target lighting image, LumiNet synthesizes a relit version of the source scene that captures the target's lighting. Our approach makes two key contributions: a data curation strategy from the StyleGAN-based relighting model for our training, and a modified diffusion-based ControlNet that processes both latent intrinsic properties from the source image and latent extrinsic properties from the target image. We further improve lighting transfer through a learned adaptor (MLP) that injects the target's latent extrinsic properties via cross-attention and fine-tuning. Unlike traditional ControlNet, which generates images with conditional maps from a single scene, LumiNet processes latent representations from two different images - preserving geometry and albedo from the source while transferring lighting characteristics from the target. Experiments demonstrate that our method successfully transfers complex lighting phenomena including specular highlights and indirect illumination across scenes with varying spatial layouts and materials, outperforming existing approaches on challenging indoor scenes using only images as input.