Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models

About

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.

Ruofan Liang, Zan Gojcic, Huan Ling, Jacob Munkberg, Jon Hasselgren, Zhi-Hao Lin, Jun Gao, Alexander Keller, Nandita Vijaykumar, Sanja Fidler, Zian Wang• 2025

Related benchmarks

TaskDatasetResultRank
Image-to-image relightingMIIW cross-scene (test)
RMSE (raw)0.399
9
Forward RenderingSyntheticScenes
PSNR26
8
Forward RenderingSyntheticObjects
PSNR28.7
8
Albedo EstimationInteriorVerse 91
PSNR22.4
7
RelightingSyntheticObjects
PSNR27.5
6
Single-view inverse renderingInteriorverse (test)
Albedo PSNR21.9
6
Forward Neural RenderingCustom Rendering Dataset (30k artist-crafted and 100k synthetic frames) (test)
PSNR23.758
6
RelightingMIIW
PSNR16.81
6
Inverse RenderingSyntheticScenes video dataset (test)
Albedo PSNR26
5
Multi-view material consistency estimationHypersim (test)
Albedo RMSE0.0826
5
Showing 10 of 23 rows

Other info

Follow for update