ComfyUI  >  Nodes  >  ComfyUI-IDM-VTON [WIP]

ComfyUI Extension: ComfyUI-IDM-VTON [WIP]

Repo Name


TemryL (Account age: 866 days)
View all nodes (2)
Latest Updated
Github Stars

How to Install ComfyUI-IDM-VTON [WIP]

Install this extension via the ComfyUI Manager by searching for  ComfyUI-IDM-VTON [WIP]
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-IDM-VTON [WIP] in the search bar
After installation, click the  Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • High-speed GPU machines
  • 200+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 50+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

ComfyUI-IDM-VTON [WIP] Description

ComfyUI-IDM-VTON [WIP] is an adaptation of IDM-VTON for virtual try-on, enabling users to visualize clothing on different body types using the ComfyUI interface.

ComfyUI-IDM-VTON [WIP] Introduction

ComfyUI-IDM-VTON is an extension for ComfyUI that adapts the IDM-VTON model to provide a seamless virtual try-on experience. This extension allows AI artists to visualize how different clothing items would look on a person, using advanced image processing techniques. Whether you're a fashion designer, a digital artist, or someone interested in virtual fashion, ComfyUI-IDM-VTON can help you create realistic and engaging virtual try-on experiences.

How ComfyUI-IDM-VTON [WIP] Works

ComfyUI-IDM-VTON leverages the power of diffusion models to generate realistic images of people wearing different outfits. Here's a simplified breakdown of how it works:

  1. Image Input: You start by providing an image of a person and an image of the clothing item you want to try on.
  2. Mask Generation: The extension uses the ComfyUI Segment Anything tool to create a mask of the person in the image. This mask helps the model understand which parts of the image correspond to the person and which parts correspond to the background.
  3. DensePose Estimation: Using ComfyUI's ControlNet Auxiliary Preprocessors, the extension estimates the DensePose of the person. DensePose is a detailed mapping of the human body that helps the model understand the 3D structure of the person in the image.
  4. Virtual Try-On: The IDM-VTON model then uses this information to generate a new image where the person is wearing the selected clothing item. The result is a realistic and high-quality virtual try-on image.

ComfyUI-IDM-VTON [WIP] Features

Mask Generation

  • What It Does: Generates a mask of the person in the image.
  • Customization: You can adjust the mask generation settings to improve accuracy.
  • Example: A well-defined mask ensures that the clothing item fits perfectly on the person without blending into the background.

DensePose Estimation

  • What It Does: Provides a detailed mapping of the human body.
  • Customization: You can fine-tune the DensePose settings to better match the body structure of the person in the image.
  • Example: Accurate DensePose estimation ensures that the clothing item conforms to the body's shape and posture.

Virtual Try-On

  • What It Does: Generates a new image of the person wearing the selected clothing item.
  • Customization: You can adjust parameters like image resolution, number of inference steps, and guidance scale to achieve the desired output quality.
  • Example: Higher resolution and more inference steps result in a more detailed and realistic virtual try-on image.

ComfyUI-IDM-VTON [WIP] Models

ComfyUI-IDM-VTON uses the IDM-VTON model, which is designed specifically for virtual try-on applications. The model weights are automatically downloaded from HuggingFace when you install the extension. Here are some key points about the model:

  • IDM-VTON: This model is optimized for generating realistic images of people wearing different outfits. It uses advanced diffusion techniques to ensure high-quality results.
  • When to Use: Use this model whenever you need to create virtual try-on images for fashion design, digital art, or any other application that requires realistic clothing visualization.

Troubleshooting ComfyUI-IDM-VTON [WIP]

Common Issues and Solutions

  1. Issue: The mask generation is not accurate.
  • Solution: Adjust the mask generation settings in the ComfyUI Segment Anything tool. Ensure that the mask accurately covers the person in the image.
  1. Issue: DensePose estimation is incorrect.
  • Solution: Fine-tune the DensePose settings in the ControlNet Auxiliary Preprocessors. Make sure the DensePose mapping aligns well with the person's body structure.
  1. Issue: The virtual try-on image looks unrealistic.
  • Solution: Increase the image resolution and the number of inference steps. Adjust the guidance scale to improve the quality of the generated image.

Frequently Asked Questions

  • Q: Do I need a powerful GPU to use ComfyUI-IDM-VTON?
  • A: Yes, the current implementation requires a GPU with at least 16GB of VRAM.
  • Q: Can I use my own images for virtual try-on?
  • A: Yes, you can use any image of a person and a clothing item as input.

Learn More about ComfyUI-IDM-VTON [WIP]

To learn more about ComfyUI-IDM-VTON and get additional support, check out the following resources:

  • : Learn more about the tool used for mask generation.
  • : Understand how DensePose estimation works.
  • : Access the model weights and additional documentation.
  • Community Forums: Join the ComfyUI community forums to ask questions, share your work, and get support from other users. By exploring these resources, you can enhance your understanding of ComfyUI-IDM-VTON and make the most out of its features for your virtual try-on projects.

ComfyUI-IDM-VTON [WIP] Related Nodes


© Copyright 2024 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals.