Add model card for Light-X

This PR adds a comprehensive model card for the Light-X model. It includes:
- A link to the paper: [Light-X: Generative 4D Video Rendering with Camera and Illumination Control](https://huggingface.co/papers/2512.05115)
- Links to the project page and the GitHub repository for further details and code.
- Essential metadata such as `pipeline_tag: image-to-video` for better discoverability on the Hub.
- The `library_name: diffusers`, enabling an automated code snippet for usage, as evidenced by `_diffusers_version` in `config.json` and acknowledgements in the project.
- The `license: apache-2.0`, based on common practice for similar open-source projects.
- Sample installation and inference instructions with code snippets directly from the GitHub README.

Please review and merge if everything looks good!

README.md

@@ -0,0 +1,52 @@
+---
+license: apache-2.0
+pipeline_tag: image-to-video
+library_name: diffusers
+---
+
+# Light-X: Generative 4D Video Rendering with Camera and Illumination Control
+
+This repository contains the official model for the paper [Light-X: Generative 4D Video Rendering with Camera and Illumination Control](https://huggingface.co/papers/2512.05115).
+
+Light-X is a video generation framework that enables controllable rendering from monocular videos with both viewpoint and illumination control. It proposes a disentangled design that decouples geometry and lighting signals, and introduces Light-Syn, a degradation-based pipeline, to synthesize training pairs from in-the-wild monocular footage.
+
+Project Page: https://lightx-ai.github.io/
+Code: https://github.com/TQTQliu/Light-X
+
+<div align="center">
+    <img src="https://github.com/user-attachments/assets/2211979f-4910-4a88-bf19-19711fa07182" width="100%"/>
+</div>
+
+## Abstract Summary
+Light-X is a video generation framework that enables controllable rendering from monocular videos with both viewpoint and illumination control. It tackles the trade-off between lighting fidelity and temporal consistency by proposing a disentangled design for geometry and lighting signals, and addresses the lack of paired multi-view and multi-illumination videos through a degradation-based pipeline called Light-Syn.
+
+## Installation
+To get started with Light-X, clone the repository and set up the environment:
+
+```bash
+git clone https://github.com/TQTQliu/Light-X.git
+cd Light-X
+conda create -n lightx python=3.10
+conda activate lightx
+pip install -r requirements.txt
+```
+
+## Inference
+You can run inference using the provided `run.sh` script. All required models will be downloaded automatically.
+
+```bash
+bash run.sh
+```
+For more detailed arguments and examples on camera and relighting control, please refer to the [official GitHub repository's inference section](https://github.com/TQTQliu/Light-X#%EF%B8%8F-inference).
+
+## Citation
+If you find our work useful for your research, please consider citing our paper:
+
+```bibtex
+@article{lightx,
+  title   = {Light-X: Generative 4D Video Rendering with Camera and Illumination Control},
+  author  = {Liu, Tianqi and Chen, Zhaoxi and Huang, Zihao and Xu, Shaocong and Zhang, Saining and Ye, Chongjie and Li, Bohan and Cao, Zhiguo and Li, Wei and Zhao, Hao and Liu, Ziwei},
+  journal = {arXiv preprint arXiv:2512.05115},
+  year    = {2025}
+}
+```