This custom node for ComfyUI provides advanced background removal capabilities using the briaai/RMBG-2.0 model and BEN2: Background Erase Network. It is designed to seamlessly integrate into the ComfyUI environment, offering users a powerful tool for image processing tasks.

Node name: GFrbmg2
Author: GorillaFrame

• High-quality background removal
• Support for batch processing
• Options for inverted masks
• Three output formats: RGBA, mask, and black background

ComfyUI/
├── models/RMBG-2.0
│   └── RMBG-2.0/  # Model files
│       ├── config.json
│       ├── model.safetensors (≈1.5GB)
│       └── other files...
│   └── BEN2/  # Model files
└── custom_nodes/
    └── GF_nodes/
        ├── gfrbmg2.py  # Node implementation
        └── __init__.py  # Node initialization

1. Create the GF_nodes Folder and Download the Model:

   Navigate to your ComfyUI directory and set up the necessary folders:

   cd ComfyUI/custom_nodes
   git clone https://github.com/gorillaframeai/GF_nodes.git

• CUDA-compatible GPU (recommended)

1. Restart ComfyUI after installation
2. Find the node in "🐵 GorillaFrame/Image" category
3. Connect an image input
4. Get outputs: image_rgba (transparent background), mask, and image_black (black background)

• Original model: briaai/RMBG-2.0 Original model: briaai/RMBG-2.0 — [License: CC BY-NC 4.0] (https://creativecommons.org/licenses/by-nc/4.0/deed.en)(Creative Commons Attribution-NonCommercial 4.0 International). Note: Use only for non-commercial purposes with proper attribution.

• Node implementation: @GorillaFrame 🐵