Add model card, pipeline tag, and links
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,3 +1,15 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
pipeline_tag: image-to-image
|
| 4 |
---
|
| 5 |
+
|
| 6 |
+
# RealVDeblur: One-Step Diffusion for Generalizable Real-World Video Deblurring
|
| 7 |
+
|
| 8 |
+
This repository contains the model presented in [RealVDeblur: One-Step Diffusion for Generalizable Real-World Video Deblurring](https://huggingface.co/papers/2607.20628).
|
| 9 |
+
|
| 10 |
+
- Project page: https://rbjin.github.io/RealVDeblur
|
| 11 |
+
- Code: https://github.com/OpenImagingLab/RealVDeblur
|
| 12 |
+
|
| 13 |
+
## Overview
|
| 14 |
+
|
| 15 |
+
RealVDeblur is a one-step video diffusion model designed for robust deblurring of real-world videos under diverse capture conditions. It leverages a physically grounded blur synthesis pipeline and a distilled one-step generator for efficient long-video inference.
|