Under One Sun: Multi-Object Generative Perception of Materials and Illumination

TL;DR

MultiGP is a novel generative inverse rendering method that disentangles reflectance, texture, and illumination from a single image by leveraging scene-wide lighting consistency, enabling detailed material and lighting recovery.

Contribution

It introduces a multi-object generative perception framework with a cascaded architecture, guidance for diffusion convergence, axial attention, and a texture control network for improved inverse rendering.

Findings

Successfully recovers individual textures and reflectance from single images.

Effectively estimates common scene illumination across multiple objects.

Demonstrates superior performance in material and lighting disentanglement.

Abstract

We introduce Multi-Object Generative Perception (MultiGP), a generative inverse rendering method for stochastic sampling of all radiometric constituents -- reflectance, texture, and illumination -- underlying object appearance from a single image. Our key idea to solve this inherently ambiguous radiometric disentanglement is to leverage the fact that while their texture and reflectance may differ, objects in the same scene are all lit by the same illumination. MultiGP exploits this consensus to produce samples of reflectance, texture, and illumination from a single image of known shapes based on four key technical contributions: a cascaded end-to-end architecture that combines image-space and angular-space disentanglement; Coordinated Guidance for diffusion convergence to a single consistent illumination estimate; Axial Attention applied to facilitate ``cross-talk'' between objects of different reflectance; and a Texture Extraction ControlNet to preserve high-frequency texture details while ensuring decoupling from estimated lighting. Experimental results demonstrate that MultiGP effectively leverages the complementary spatial and frequency characteristics of multiple object appearances to recover individual texture and reflectance as well as the common illumination.