May 2023 saw Adobe integrate its Firefly AI image technology into the Photoshop (beta) app with Generative Fill. This allowed Photoshop users to add, expand, or remove content from images non-destructively through the use of simple text prompts. To date, this feature has been instrumental in generating over 900 million images and has opened up a world of possibility for users, allowing for an imaginative expansion of dimensions, pixel transformations, and more intuitive and efficient editing.
Recently, Adobe has announced that the Firefly-powered features in Photoshop (beta) now support text prompts in over 100 languages. The company says this marks a significant step towards inclusivity, enabling users around the world to express their creative vision in their preferred language.
Adobe is also enhancing Photoshop’s Generative AI capabilities with the introduction of Generative Expand. This new workflow is dedicated to expanding images beyond their original aspect ratios. Like Generative Fill, Generative Expand is powered by the Firefly-based technology, and it allows users to expand and resize any image seamlessly using the Crop tool. With a simple click on “Generate”, the AI fills the empty space with content that blends naturally with the existing image, creating a cohesive and appealing result.
According to Adobe, “If you’ve ever faced challenges with your subject being cut off, your image not aligning with the desired aspect ratio, or objects in focus misaligned with other parts of the image, Generative Expand offers a solution. The tool enables you to expand your canvas and bring your imagination to life on the image.”
The process begins with selecting the Crop tool and dragging beyond the image’s original canvas to reach the desired aspect ratio. Generated content can be added with or without a text prompt. Without a prompt, by clicking “Generate” in the Contextual Task Bar, Photoshop fills in the newly expanded canvas with generated content that blends seamlessly with the existing image.
When using a prompt, the image is expanded to include the content entered into the prompt. Once you select your favourite variation, the expanded image is added non-destructively in a new Generative Layer.
“Thanks to Generative Expand, creatives can now spend less time editing and more time experimenting and adapting their images to meet their unique creative needs,” says Adobe.
Adobe says it is planning to announce more generative AI developments in 3Q 2023.