Raster Layers and Initial Images

Raster Layer

Using an Initial Image as a Raster Layer allows you to inspire the generation process with an initial drawing or image, which preserves the original image's rough structure, colors, and layout, while using AI to reimagine new content with your input prompt.

Using your Initial Image as a Raster Layer

To run an image-to-image process on an existing image, simply drag the image into the canvas and select ‘Create Raster Layer.’ Depending on the size of your image, you may then need to resize your image to the dimensions you want to generate at, or use the the ‘Fit BBox to Layers’ button in the upper right of the Canvas panel to adjust your bounding box. You’ll want to generate new images within the ranges specified by your selected model.

To run an image-to-image process and save each of those as unique images to the gallery, select the "Save to Gallery" toggle next to the Invoke button. To run image-to-image and stage that as a work in progress on the Canvas which you can review and edit, select the "Save to Canvas" toggle. You can then later save that to the gallery once you're satisfied with the final output.

Controlling Image-to-Image transformations

You can control how closely the system should stay as it generates the image by controlling the Denoising Strength, which ranges from 0.0 (keep the original intact) to 1.0 (ignore the original completely). The default is 0.75, and values between 0.25 and 0.90 often give interesting results. A decent guideline for using this is:

• .35 will produce very minor variations on the image.
• .55 will produce very moderate variations on the image.
• .75 or more will produce significant variations on the image.

How Does It Work?

The main difference between using Image to Image and a regular Text to Image generation is the starting point of the generation process.

While Text to Image starts with pure noise and refines it over the requested number of steps, Image to Image skips some of these earlier steps (based on your Denoising Strength) and uses your initial image as the starting point. Depending on the set strength, your image will be inserted into the sequence and get “finished” by the denoising process - A higher strength means more denoising will be done, and more details added by the AI.

See below for an example of a normal Text to Image generation, and how the generated output changes when an Image is supplied with a 0.6 strength.