---
title: diffuEraser
canonical_url: "https://www.modelscope.cn/models/xingzi/diffuEraser"
md_url: "https://www.modelscope.cn/models/xingzi/diffuEraser.md"
repository: xingzi/diffuEraser
chinese_name: "视频擦除扩散模型-DiffuEraser"
last_updated: 2025-04-08
license: "Apache License 2.0"
pipeline_tag: video-editing
tasks:
  - video-editing
model_type:
  - "stable diffusion"
parameters: 2.2B
tensor_type:
  - F32
library_name:
  - safetensors
  - diffusers
  - pytorch
frameworks:
  - Pytorch
domain:
  - cv
downloads: 1673
stars: 8
tags:
  - "video inpainting"
  - "diffusion model"
---

# diffuEraser

> diffuEraser - xingzi 在 ModelScope 开源的模型。基于扩散模型的视频擦除模型

xingzi/diffuEraser 是 ModelScope 魔搭社区上的 2.2B 参数video-editing模型，采用 Apache License 2.0 许可。

- **Repository**: xingzi/diffuEraser
- **License**: Apache License 2.0
- **Tasks**: video-editing
- **Parameters**: 2.2B
- **Tags**: video inpainting, diffusion model
- **Downloads**: 1673
- **Stars**: 8
- **Last updated**: 2025-04-08

Source: https://www.modelscope.cn/models/xingzi/diffuEraser

---

# DiffuEraser: A Diffusion Model for Video Inpainting

DiffuEraser is a diffusion model for video inpainting, which outperforms state-of-the-art model Propainter in both content completeness and temporal consistency while maintaining acceptable efficiency.

## Model Sources

- **Project Page:** https://lixiaowen-xw.github.io/DiffuEraser-page
- **Repository:** https://github.com/lixiaowen-xw/DiffuEraser.git
- **Paper:** https://arxiv.org/abs/2501.10018
- **ModelScope Demo:** https://www.modelscope.cn/studios/iic/DiffuEraser

## Quick Start
#### 您可以通过如下git clone命令，或者ModelScope SDK来下载模型
SDK下载
```bash
#安装ModelScope
pip install modelscope
```
```python
#SDK模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('xingzi/diffuEraser')
```
Git下载
```
#Git模型下载
git clone https://www.modelscope.cn/xingzi/diffuEraser.git
```

## Citation 

**BibTeX:**

```bibtex
@misc{li2025diffueraserdiffusionmodelvideo,
  title={DiffuEraser: A Diffusion Model for Video Inpainting}, 
  author={Xiaowen Li and Haolan Xue and Peiran Ren and Liefeng Bo},
  year={2025},
  eprint={2501.10018},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2501.10018}, 
}
```


## Model Card Authors

Xiaowen Li / xwlxw@qq.com / lxw262398@alibaba-inc.com