README.md

---
license: other
task_categories:
- image-to-image
tags:
- brdfusion
- inverse-rendering
- relighting
- urban-scenes
- synthetic-data
- waymo-open-dataset
---

# BRDFusion Preprocessed Data

This repository contains the preprocessed evaluation and training data for the paper **BRDFusion: Physics Meets Generation for Urban Scene Inverse Rendering**.

[Project Page](https://shigon255.github.io/brdfusion-page/) | [GitHub](https://github.com/shigon255/BRDFusion) | [arXiv](https://arxiv.org/abs/2606.17049)

BRDFusion combines physics-based inverse rendering with generative modeling for high-quality urban scene relighting. It decomposes urban videos into geometry, materials, and HDR lighting for novel-view rendering, relighting, and scene-edit applications.

## Citation

```bibtex
@misc{liu2026brdfusionphysicsmeetsgeneration,
  title={BRDFusion: Physics Meets Generation for Urban Scene Inverse Rendering}, 
  author={Yi-Ruei Liu and Jie-Ying Lee and Zheng-Hui Huang and Yu-Lun Liu and Chih-Hao Lin},
  year={2026},
  eprint={2606.17049},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2606.17049}, 
}
```