DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models

TL;DR

DiffusionRenderer is a neural framework that jointly performs inverse and forward rendering from videos using diffusion models, enabling realistic editing and relighting without explicit scene geometry or lighting data.

Contribution

It introduces a unified neural approach leveraging video diffusion priors for inverse and forward rendering, outperforming existing methods and enabling practical scene editing from videos.

Findings

Accurately estimates G-buffers from real-world videos.

Generates photorealistic images from G-buffers.

Enables relighting, material editing, and object insertion from a single video.

Abstract

Understanding and modeling lighting effects are fundamental tasks in computer vision and graphics. Classic physically-based rendering (PBR) accurately simulates the light transport, but relies on precise scene representations--explicit 3D geometry, high-quality material properties, and lighting conditions--that are often impractical to obtain in real-world scenarios. Therefore, we introduce DiffusionRenderer, a neural approach that addresses the dual problem of inverse and forward rendering within a holistic framework. Leveraging powerful video diffusion model priors, the inverse rendering model accurately estimates G-buffers from real-world videos, providing an interface for image editing tasks, and training data for the rendering model. Conversely, our rendering model generates photorealistic images from G-buffers without explicit light transport simulation. Experiments demonstrate that DiffusionRenderer effectively approximates inverse and forwards rendering, consistently outperforming the state-of-the-art. Our model enables practical applications from a single video input--including relighting, material editing, and realistic object insertion.