The new Adobe Photoshop gets an in-app image generator, major Generative Fill upgrades

Hyperrealistic cordyceps mushrooms in pupil prompt output using Firefly Image 3 Model. 

Adobe

When Adobe unveiled Generative Fill for Photoshop last May, the internet went wild because of its ability to take photo editing to new heights. With Adobe’s latest Photoshop update, the application is not only getting a new in-app text-to-image generator, but also Generative Fill upgrades.

On Tuesday at Adobe Max London, the company unveiled the beta of its new Photoshop, which boasts significant upgrades including an in-app image generator powered by an all-new Firefly image model, a more advanced Generative Fill, and other new features and tools to optimize photo editing. 

Also: The best AI image generators of 2024: Tested and reviewed

“This massive new update to Photoshop takes creativity to unprecedented levels – whether you are a professional or brand new to Photoshop,” said Ashley Still, SVP of Creative Product Group at Adobe.

The upgraded Photoshop makes it easier for users to add or tweak elements to their projects; new tools include the Generate Image feature, which enables users to generate images from text prompts without leaving the application. 

Adobe

The upgraded Generative Fill enables users to upload Reference Images to generate new elements for their project that best fit the reference picture’s style, as seen below.

Adobe

Similarly, the Generate Similiar feature enables users to import images and generate variations that they can add to their projects, as demonstrated in the video below. 

Adobe

The new Generate Background feature enables users to replace or create a background of an image with a simple text prompt, as seen in the photo below. Lastly, with Enhance Detail, users can improve the clarity and sharpness of an image, according to the release. 

READ MORE  16 Best Tablets (2023): For Work, Drawing, Play, Kids, and Seniors

Adobe

All of the above generation capabilities are powered by the Firefly Image 3 Foundation Model, a new and improved version of Firefly that Adobe shares can deliver better quality renditions, higher prompt accuracy, clear text displays, and more customization options. 

Also: Google releases two new free resources to help you optimize your AI prompts

The viral Generative Fill and Generative Expand features have also been supercharged with Firefly Image 3 to produce even better results.

The new Firefly Image 3 model is accessible in beta via the Firefly web application on the Firefly website, where users will also be able to access the new Generative Expand and Generative Fill Features. 

Adobe reiterates that Adobe Firefly was designed for commercial use, being trained on licensed content that doesn’t infringe on copyright and other intellectual property. 

The company also notes that it uses a continuous review approach to block and remove content that violates Adobe’s policies, likely a nod to the controversy it recently faced regarding the content used to train its image model. 

In addition to the new Firefly features, Photoshop also got other new, useful tools such as Adjustment Brush; which allows users to apply “non-destructive adjustments” to parts of their images; Improved Font Browser, which gives users real-time access to Adobe’s font library without leaving Photoshop; and Adjustment Presets, which allows users to apply a filter with one click. 

Leave a Comment