Expand the content of an image using outpainting with Imagen

Editing with Imagen is only supported if you're using the Vertex AI Gemini API . It's also currently only supported for Android and Flutter apps. Support for other platforms is coming later in the year.

This page describes how to use outpainting using Imagen to expand the content of an image beyond its original bordersusing the Firebase AI Logic  SDKs.

Outpainting is a type of mask-based editing . A mask is a digital overlay defining the specific area you want to edit.

How it works: You provide an original image and a corresponding masked image — either auto-generated or provided by you — that defines a mask of the new, expanded area. You can also optionally provide a text prompt describing what you want in the expanded area, or the model can intelligently decide what will logically continue the existing scene. The model generates the new content and fills in the masked area.

For example, you can change an image's aspect ratio or add more background context.

arrow_downward Jump to code

Before you begin

If you haven't already, complete the getting started guide , which describes how to set up your Firebase project, connect your app to Firebase, add the SDK, initialize the backend service for your chosen API provider, and create an ImagenModel instance.

Models that support this capability

Imagen offers image editing through its capability model:

• imagen-3.0-capability-001

Note that for Imagen models, the global location is not supported.

Expand content of an image

The following sample shows how to expand an image beyond its original borders — using a mask defined in an image that you provide. You provide the original image, a text prompt, and the masked image. Note the following about the original and masked image:

• The masked image must have the pixel dimensions of the targeted size of the final outpainted image.

• The original image must include additional padding to match the pixel dimensions of the masked image.

Providing a text prompt is optional if you want the model to intelligently decide what will logically continue the existing scene. If you want specific content within the expanded area, you need to specify that in a text prompt.

Best practices and limitations

We recommend dilating the mask when editing an image. This can help smooth the borders of an edit and make it seem more convincing. Generally, a dilation value of 1% or 2% is recommended ( 0.01 or 0.02 ).

Give feedback about your experience with Firebase AI Logic