ComfyUI-RMBG

A comprehensive repository designed for users of ComfyUI looking to enhance their image processing capabilities with advanced background removal, object, face, clothes, and fashion segmentation. This repository leverages multiple robust models such as RMBG-2.0, INSPYRENET, BEN, BEN2, BiRefNet-HR, SAM, and GroundingDINO to provide a wide array of image processing functionalities.

Installation

Method 1: ComfyUI-Manager

1. Search for ComfyUI-RMBG in the ComfyUI-Manager.

2. Install the package.

3. Execute the following command to install the required dependencies:

   ./ComfyUI/python_embeded/python -m pip install -r requirements.txt
   

Method 2: Clone This Repository

1. Navigate to your ComfyUI custom nodes directory:

   cd ComfyUI/custom_nodes
   

2. Clone the repository:

   git clone https://github.com/1038lab/ComfyUI-RMBG
   

3. Install the required dependencies:

   ./ComfyUI/python_embeded/python -m pip install -r requirements.txt
   

Method 3: ComfyCLI

1. Ensure comfy-cli is installed using pip install comfy-cli.

2. Install ComfyUI with comfy install if not already installed.

3. Install ComfyUI-RMBG using:

   comfy node install ComfyUI-RMBG
   

4. Install the required dependencies:

   ./ComfyUI/python_embeded/python -m pip install -r requirements.txt
   

Method 4: Manually Download Models

• Models are automatically downloaded to ComfyUI/models/RMBG/ when first utilizing the custom node.
• Manual download links are available for specific models like RMBG-2.0, INSPYRENET, and others at the respective HuggingFace pages.

Repository Purpose

The repository is designed to facilitate advanced image manipulation within ComfyUI, offering tools for background removal and segmentation across different objects and features such as faces, clothes, and accessories. The custom nodes integrate seamlessly into ComfyUI workflows, providing high precision and flexibility with support for batch processing and multiple customization options.

Provided Nodes

This repository includes a variety of nodes tailored to specific tasks:

• BiRefNetRMBG
• BodySegment
• ClothesSegment
• FaceSegment
• FashionSegmentAccessories
• FashionSegmentClothing
• AILab_LoadImage
• AILab_LoadImageSimple
• AILab_LoadImageAdvanced
• AILab_Preview
• AILab_MaskOverlay
• AILab_ImagePreview
• AILab_MaskPreview
• AILab_ImageMaskConvert
• AILab_MaskEnhancer
• AILab_MaskCombiner
• AILab_ImageCombiner
• AILab_MaskExtractor
• AILab_ImageStitch
• AILab_ImageCrop
• AILab_ICLoRAConcat
• AILab_CropObject
• AILab_ImageCompare
• AILab_ColorInput
• AILab_ImageMaskResize
• AILab_ReferenceLatentMask
• AILab_LamaRemover
• RMBG
• Segment
• SegmentV2

Special Features and Capabilities

• Background Removal and Segmentation: Provides advanced algorithms for accurate background removal and object segmentation with models like RMBG-2.0, SAM, and GroundingDINO.
• High-Resolution Support: Fully supports high-resolution image processing, suitable for complex scenes with fine details.
• Internationalization: Enhances accessibility with multilingual support, allowing dynamic language switching within the user interface.
• Optimization Options: Custom nodes with optimization settings to fine-tune performance and quality.
• Innovative Models: Utilizes state-of-the-art models for tasks such as semantic segmentation and matting.

Application in ComfyUI Workflows

By integrating ComfyUI-RMBG into workflows, users can streamline the process of background removal and segmentation to enhance their image processing tasks. The custom nodes support various segmentation modes and resolutions, making them adaptable for different needs, from small-scale images to complex, detailed scenes. The nodes also provide options for batch processing, enabling efficient handling of large volumes of images.

Overall, ComfyUI-RMBG proves to be a vital tool for users looking to expand their image manipulation capabilities within the ComfyUI environment, enhancing both workflow efficiency and output quality.