ComfyUI Image Processing Guide: Img2Img Tutorial

The Img2Img feature in ComfyUI allows for image transformation. Here’s a quick guide on how to use it:

  1. Preparing Your Images:
    • Ensure your target images are placed in the input folder of ComfyUI.
  2. Setting up the Workflow:
    • Navigate to ComfyUI and select the examples. From there, opt to load the provided images to access the full workflow.
  3. Utilizing Img2Img:
    • With Img2Img, you’ll initiate by choosing your desired image. This image is then converted in the latent space using the VAE.
    • The key variable here is the ‘denoise’ setting. By adjusting this, you influence the level of noise introduced to the image during processing. A value below 1.0 is ideal. For instance, setting it at 0.87 means less noise is added, retaining more of the image’s original characteristics.
  4. Executing the Workflow:
    • In this streamlined workflow, the process mirrors the default txt2img, with two variations:
      • The ‘denoise’ is adjusted to 0.87.
      • Instead of an empty placeholder, a selected image from your input folder is utilized for sampling.

In following this guide, you’ll be able to transform and refine images efficiently using ComfyUI’s Img2Img feature.


Leave a Reply