What are Layers?

Layers are a powerful set of tools enabling you to control and guide the image generation process to produce a desired outcome.

Types of Layers

Global and Regional Reference Images

Global and Regional Reference Image layers use Image Prompt (IP) Adapters to inspire a new image with the content of an input image. You can use add both Global and Regional Reference Images as layers on the ‘Layers’ tab.

Global References Image layers influence the entire bounding box (regardless of position), while Regional Reference Images can be used to influence specific parts of an image on the Control Canvas and stay fixed to the area that you select using the brush.

You can learn more about using Global and Regional Image layers here.

Inpaint Mask

The Inpaint Mask layer allows you to specify a region that will be modified for generation, while preserving the rest of your raster layer data. Results will be blended seamlessly with the surrounding content, and any newly generated content will be added as a new layer.

The Denoising Strength that you select will dictate how much change you want the AI model to generate in the selected region. At very high denoising strengths, the newly generated content will be very different from your original image, and at low denoising strengths, it will only make minor changes.

At high denoising strengths, it can be helpful to also utilize a control layer so that the composition and structure of your image matches, despite the significant changes being applied.

You can learn more about using Inpaint Mask layers here.

Regional Guidance

Regional Guidance layers allow you more fine-tuned control over the prompt information used to guide the generation process. With the ability to use Positive, Negative, and Reference Image prompts, you can brush specific areas to instruct the generation with content it should use for that area. These are supplemental in nature, and should typically align with your general/global image prompt on the left panel, for the highest quality images.

You can read more about text-based Regional Guidance here and image based Regional Guidance here.

Control Layer

With Control Layers, you can provide structural control over the output of your image generations, with a number of different ways to instruct the system using visual representations like sketches, edge maps, and depth renderings.

Control Layers work by analyzing an input image (sometimes requiring you to utilize the “filter” option if utilizing an image that needs to be in a specific format) to identify relevant information that can be interpreted by the Control Layer model you have selected. This can be used to control the composition and other aspects of the image to better achieve a specific result.

You can learn more about using Control Layers here.

Raster Layer

A Raster Layer is the image content of your canvas, similar to other Image Editing solutions. When included in your bounding box, these images serve as the base image content to start your creative process, and leverage the Denoising Strength that you have configured.

This 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 based on your denoising strength.

You can learn more about utilizing Raster Layers, including as an initial image, here.

Accessing and Managing Layers

There are two ways to add layers in the Control Canvas.

First, you can add all layer types by clicking the ‘+’ icon at the top right corner of the ‘Layers’ tab. Select your desired layer type from the dropdown menu, then configure the layer settings once the layer has been added to the list of layers in your ‘Layers’ tab.

Alternatively, you can select an image or asset from the ‘Gallery’ tab and drag to the portion of the canvas that corresponds to the desired layer type. This is a great way to add Raster, Control, and Reference Image layers - It will automatically create the layer, and use the dragged image as the content for that layer.

To manage the layers you have deployed on the Canvas, you can view the active layers in the ‘Layers’ tab. This tab can be used to edit the settings of each layer individually. You can show, hide, and delete layers as you iterate on your work.