ComfyUI Extension
ComfyUI_ObjectClear
ObjectClear:Complete Object Removal via Object-Effect Attention,you can try it in ComfyUI
smthemex/ComfyUI_ObjectClear
Nodes4
On cloudLocal install
CategoryObjectClear
Stars34
Updated4 days ago
Nodes (4)
Readme
ComfyUI_ObjectClear
ObjectClear:Complete Object Removal via Object-Effect Attention,you can try it in ComfyUI
Update
- 调用agent重新封装此方法 // call agent to encapsulate this method again
1 . Installation /安装
In the ./ComfyUI /custom_nodes directory, run the following:
git clone https://github.com/smthemex/ComfyUI_ObjectClear.git
2 . Requirements
pip install -r requirements.txt
3 . Model/模型
- jixin0101/ObjectClear download files as below /下载下方列出的模型:
├── comfyUI/models/checkpoints/
| ├──diffusion_pytorch_model.safetensors #rename it
├── vae/
| ├──sdxl.vae.safetensors
├── clip_vision/
| ├── clip-vit-large-patch14.safetensors
├── clip/
| ├── postfuse_module.safetensors
4 . Example

5 .Tips/使用说明
- 支持单图或多图,遮罩合并用插件带的batch节点,512是垫图最小短边尺寸(会自动裁切),模型用512*512训练的,按理此尺寸是最好的。
- Support single or multiple images, use batch nodes with plugins for mask merging, 512 is the input image ‘s minimum short edge size, and the model is trained with 512 * 512, which is theoretically the best size.
6. Citation
@InProceedings{zhao2025ObjectClear,
title = {{ObjectClear}: Complete Object Removal via Object-Effect Attention},
author = {Zhao, Jixin and Zhou, Shangchen and Wang, Zhouxia and Yang, Peiqing and Loy, Chen Change},
booktitle = {arXiv preprint arXiv:2505.22636},
year = {2025}
}