Add pipeline tag, paper link and sample usage

Hi! I'm Niels from the Hugging Face team.

This PR improves the model card for the RealRestorer Degradation Models by:
- Adding the `pipeline_tag: image-to-image` to the metadata for better discoverability.
- Adding a link to the paper [RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models](https://huggingface.co/papers/2603.25502).
- Adding a sample usage section for the degradation pipeline based on the GitHub README.

Please let me know if you have any questions!

README.md

@@ -1,27 +1,35 @@
 ---
 license: apache-2.0
+pipeline_tag: image-to-image
 ---
 
 # RealRestorer Degradation Models
 
-This repository provides the degradation models used in the degradation pipeline of [RealRestorer](https://github.com/yfyang007/RealRestorer).
+This repository provides the degradation models used in the degradation pipeline of **RealRestorer**, as presented in the paper [RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models](https://huggingface.co/papers/2603.25502).
 
 In particular, these models are used for **moire synthesis** and **specular reflection synthesis** in the RealRestorer degradation pipeline, which are designed to generate realistic degradations for training data construction and controlled degradation simulation.
 
-## Usage
+## Links
 
-Please refer to the main RealRestorer repository for the complete pipeline, code, and usage instructions:
+- **Paper**: [RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models](https://huggingface.co/papers/2603.25502)
+- **Project Page**: [https://yfyang007.github.io/RealRestorer/](https://yfyang007.github.io/RealRestorer/)
+- **Code**: [https://github.com/yfyang007/RealRestorer](https://github.com/yfyang007/RealRestorer)
+- **RealRestorer Model**: [Hugging Face](https://huggingface.co/RealRestorer/RealRestorer)
+- **RealIR-Bench Dataset**: [Hugging Face](https://huggingface.co/datasets/RealRestorer/RealIR-Bench)
 
-- **RealRestorer**: https://github.com/yfyang007/RealRestorer
+## Usage
 
 These degradation models are intended to be used together with the degradation pipeline in RealRestorer rather than as a standalone image restoration model.
 
-## Related Links
-
-- **Project Page**: https://yfyang007.github.io/RealRestorer/
-- **RealRestorer Model**: https://huggingface.co/RealRestorer/RealRestorer
-- **RealIR-Bench**: https://huggingface.co/datasets/RealRestorer/RealIR-Bench
+To synthesize degradations (e.g., reflection) using these models, you can use the generation script provided in the [official repository](https://github.com/yfyang007/RealRestorer):
 
+```bash
+python3 infer_degradation.py \
+  --image /path/to/input.png \
+  --degradation reflection \
+  --output /path/to/degraded.png \
+  --reflection_ckpt_path /path/to/130_net_G.pth
+```
 
 ## Citation
 
@@ -55,4 +63,4 @@ If you find RealRestorer useful in your research, please star and cite:
   pages={3771--3779},
   year={2019}
 }
-```
+```