batchImg-rembg

Rembg(Remove background) of image sequence for ComfyUI

Installation

1. Clone to your custom_nodes folder in ComfyUI:

git clone https://github.com/Mamaaaamooooo/batchImg-rembg-ComfyUI-nodes.git

2. Install rembg[gpu] (recommended) or rembg, depending on GPU support, to your ComfyUI virtual environment. E.g.:

pip install rembg[gpu]
pip install tqdm

batchImg-rembg workflows will often make use of these helpful node packs:

---

• ComfyUI-AnimateDiff-Evolved for img2vid.

• ComfyUI-VideoHelperSuite for loading videos, combining images into videos.

• comfyui_controlnet_aux for preprocessing original images to depth, lineart, openpose images.
  

Workflow

for example,

Optional Models(Choose according to your work, downloaded automatically)

---

All models are downloaded and saved in the user home folder in the .u2net directory.

The available models are:

Rembg Model Name

Name	Description	Link
u2net(default)	A pre-trained model for general use cases.	download, source
u2netp	A lightweight version of u2net model.	download, source
u2net_human_seg	A pre-trained model for human segmentation.	download, source
u2net_cloth_seg	A pre-trained model for Cloths Parsing from human portrait. Here clothes are parsed into 3 category: Upper body, Lower body and Full body.	download, source
silueta	Same as u2net but the size is reduced to 43Mb.	download, source
isnet-general-use	A new pre-trained model for general use cases.	download, source
isnet-anime	A high-accuracy segmentation for anime character.	download, source
sam(not recommended, not easy to use)	A pre-trained model for any use cases.	download encoder, download decoder, source

Example

Original Image	LineArt before Rembg	LineArt after Rembg

Acknowledgements

Thanks to rembg-comfyui-node for insight.
Thanks to rembg-comfyui-node-better for modifying repo.