Run ComfyUI Easily with InstaSD

Skip the complex setup and run ComfyUI online. InstaSD helps creative professionals build workflows and deploy them to the world:

  • One-click deployment
  • Any model, any node
  • Powerful GPUs for rapid iteration
Get Started

Available Nodes

easy imageRemBg

ComfyUI Node Documentation: Easy Image Remove Background

Overview

The "easy imageRemBg" node is part of the ComfyUI-Easy-Use project, a tool that enhances workflow efficiency in ComfyUI. This node specializes in removing background elements from images using several selectable algorithms. It is designed to streamline the process of obtaining transparent-background images, which can be crucial for tasks in design, photography, and content creation.

Functionality

This node removes the background from an image by applying different segmentation models. It supports multiple algorithms to cater to various user needs, providing flexibility and precision in background removal.

Input Parameters

  • Images: This input expects one or more images from which the background will be removed.
  • Removal Mode (rem_mode): Users can choose the algorithm to use for background removal. Options include:
    • RMBG-1.4: A fast and efficient model typically used for general-purpose background removal.
    • RMBG-2.0: An advanced model for higher precision, particularly useful for more complex images.
    • Inspyrenet: A method focused on removing backgrounds for creative and artistic image styles.
    • BEN2: Offers refined foreground separation for challenging scenarios.
  • Image Output Options:
    • Hide: The output image will not be shown within the UI.
    • Preview: Displays the output image as a preview within the UI.
    • Save: Saves the output image to the specified directory.
    • Hide/Save: The image will be saved but not displayed.
  • Save Prefix (save_prefix): A prefix to be used when saving the output file names.
  • Torchscript JIT: An optional toggle to enable Just-In-Time (JIT) compilation for faster processing in some models.
  • Add Background: Allows the addition of a solid color background post-removal. Options include "none", "white", or "black".
  • Refine Foreground: A boolean option to enhance the quality of the foreground, useful in scenarios involving complex subject edges.
  • Prompt and Extra PNG Info (hidden): Metadata used internally by ComfyUI.

Outputs

  • Image: This is the processed image with the background removed, which can be further utilized in other nodes or saved for reviewing and editing.
  • Mask: A binary mask that delineates the removed background. This can be useful for further refining the image or combining it with other effects.

Usage in ComfyUI Workflows

In a typical workflow, the "easy imageRemBg" node can be inserted wherever background removal is needed. It can be used:

  • As a preprocessing step: Prepare images by removing backgrounds before feeding them into other image processing or analysis nodes.
  • For generating assets: Create subject-focused images for use in digital art, advertisements, or content creation without backgrounds.
  • In complex pipelines: Post-process images in a multi-step pipeline where background removal is a critical task.

Special Features and Considerations

  • Multiple Algorithms: Select from multiple models to ensure the best results for various kinds of images.
  • Efficient Processing: Designed to handle the task of background removal efficiently, even for high-resolution images.
  • Customizability: The node allows flexibility with additional backgrounds, output options, and algorithm selection, making it suitable for both quick and detailed tasks.
  • Ease of Integration: Can be seamlessly integrated into ComfyUI workflows thanks to its straightforward input-output design.

This versatile node offers optimal background removal features that enable creative professionals to tailor the outcomes to fit their specific project needs.

Want to try ComfyUI-Easy-Use without installing anything? Run ComfyUI online on InstaSD's cloud GPUs — no setup required.