ComfyUI_DiffuEraser

★ 258

视频修复视频Inpainting扩散模型ComfyUI插件

将DiffuEraser扩散视频修复模型集成到ComfyUI，便于在可视化节点流程中对视频序列进行遮挡移除与修复。

💡 在ComfyUI中对需移除或修复的视频对象进行帧级修补。

🍴 19 Forks💻 Python🔄 2025-12-02

🔗 GitHub 原文

📦

网盘下载

复制链接后前往夸克网盘下载

https://pan.quark.cn/s/c1eafc754fbb

📦 requirements.txt

torch
torchvision
torchaudio
diffusers
accelerate
opencv-python
imageio
#matplotlib
transformers
einops
#datasets
#numpy==1.26.4
#pillow==10.4.0
#tqdm==4.66.4
#urllib3==2.2.2
#zipp==3.19.2
peft
#scipy==1.13.1
#av==14.0.1

📄 README

ComfyUI_DiffuEraser

DiffuEraser is a diffusion model for video Inpainting, you can use it in ComfyUI

Update

使用官方推荐的vae文件，clip-l改成comfyUI 默认的，由此剔除掉sd15底模；

对于水印，mask_dilation_iter（遮罩膨胀系数）应适当调低，比如2，常规使用Propainter的采样也就够了（最大边是960，应用了再超分吧）

use cofmyUI v3 mode,fix bugs,add new diffuser support,you can run 1280*720 (12GVRAM) now

修复不少bug，现在12G也能跑1280*720，DiffuEraser的sample 节点的 blend支持2种输出，关闭为降低闪烁，开启为使用合成，避免loop循环的反复加载模型

1. Installation

In the ./ComfyUI /custom_nodes directory, run the following:

git clone https://github.com/smthemex/ComfyUI_DiffuEraser.git

2. Requirements

no need, because it’s base in sd1.5 ,Perhaps someone may be missing the library.没什么特殊的库,懒得删了

pip install -r requirements.txt

3. Models

vae links

clip-l， comfyUI normal

pcm 1.5 lora address pcm_sd15_smallcfg_2step_converted.safetensors #example

ProPainter address # below example

unet and brushnet address # below example

--  ComfyUI/models/vae
    |-- sd-vae-ft-mse.safetensors #vae
--  ComfyUI/models/clip
    |-- clip_l.safetensors # comfyUI normal
--  ComfyUI/models/DiffuEraservae
     |--brushnet
        |-- config.json
        |-- diffusion_pytorch_model.safetensors
     |--unet_main
        |-- config.json
        |-- diffusion_pytorch_model.safetensors
     |--propainter
        |-- ProPainter.pth
        |-- raft-things.pth
        |-- recurrent_flow_completion.pth

If use video to mask #可以用RMBG或者BiRefNet模型脱底

-- any/path/briaai/RMBG-2.0   # or auto download 
        |--config.json
        |--model.safetensors
        |--birefnet.py
        |--BiRefNet_config.py
Or
-- any/path/ZhengPeng7/BiRefNet   # or auto download 
        |--config.json
        |--model.safetensors
        |--birefnet.py
        |--BiRefNet_config.py
        |--handler.py

4 Example

use single mask

5.Citation

@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}, 
}

@inproceedings{zhou2023propainter,
   title={{ProPainter}: Improving Propagation and Transformer for Video Inpainting},
   author={Zhou, Shangchen and Li, Chongyi and Chan, Kelvin C.K and Loy, Chen Change},
   booktitle={Proceedings of IEEE International Conference on Computer Vision (ICCV)},
   year={2023}
}

@misc{ju2024brushnet,
  title={BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion}, 
  author={Xuan Ju and Xian Liu and Xintao Wang and Yuxuan Bian and Ying Shan and Qiang Xu},
  year={2024},
  eprint={2403.06976},
  archivePrefix={arXiv},
  primaryClass={cs.CV}
}

@article{BiRefNet,
  title={Bilateral Reference for High-Resolution Dichotomous Image Segmentation},
  author={Zheng, Peng and Gao, Dehong and Fan, Deng-Ping and Liu, Li and Laaksonen, Jorma and Ouyang, Wanli and Sebe, Nicu},
  journal={CAAI Artificial Intelligence Research},
  year={2024}
}