NDJIR: Neural Direct and Joint Inverse Rendering for Geometry, Lights, and Materials of Real Object

TL;DR

NDJIR introduces a neural inverse rendering approach that directly models the rendering equation to accurately decompose geometry, lighting, and materials from multi-view images of real objects, enabling seamless mesh and material export.

Contribution

It presents a novel neural inverse rendering method that directly addresses the rendering equation and jointly decomposes geometry, lights, and materials using volume rendering and Bayesian priors.

Findings

Effective decomposition of geometry, lights, and materials on real objects.

Seamless export of meshes and materials to DCC tools.

Improved accuracy in photogrammetric inverse rendering.

Abstract

The goal of inverse rendering is to decompose geometry, lights, and materials given pose multi-view images. To achieve this goal, we propose neural direct and joint inverse rendering, NDJIR. Different from prior works which relies on some approximations of the rendering equation, NDJIR directly addresses the integrals in the rendering equation and jointly decomposes geometry: signed distance function, lights: environment and implicit lights, materials: base color, roughness, specular reflectance using the powerful and flexible volume rendering framework, voxel grid feature, and Bayesian prior. Our method directly uses the physically-based rendering, so we can seamlessly export an extracted mesh with materials to DCC tools and show material conversion examples. We perform intensive experiments to show that our proposed method can decompose semantically well for real object in photogrammetric setting and what factors contribute towards accurate inverse rendering.