Adobe has introduced an experimental feature set called AI Playground to its Project Indigo camera app, a tool previously known for its manual controls and natural, SLR-like image processing. The update, which is currently rolling out to a small percentage of users, integrates generative AI capabilities directly into the mobile photography workflow.
The new AI Playground includes four primary sections: Object Editing, Styles, Photo Guidance, and Custom Edit. Object Editing allows users to remove distractors or apply background defocusing for a shallow depth-of-field effect. The Styles menu provides buttons to re-render images using various artistic techniques, while the Custom Edit feature enables users to apply arbitrary text prompts to their photos.
The Photo Guidance tool functions as an automated critic, analyzing captured images to provide suggestions for future reshoots or specific editing adjustments. Adobe notes that these features are currently limited to images captured within the Project Indigo app itself, meaning imported photos from other sources are not compatible with the AI tools.
Under the hood, the app utilizes Google’s Nano Banana model to power these generative tasks. Because this model outputs images in Standard Dynamic Range (SDR), Adobe has implemented a custom machine learning model to post-process these images and restore High Dynamic Range (HDR) quality. Digital Trends reports that the tool requires an active internet connection and currently caps edited output at 2K resolution.
Adobe has stated that the AI Playground is free to use for the duration of the limited test, which is expected to last a few weeks. The company is not requiring a sign-on for the experiment and claims that analytics gathered are anonymous, with no prompts or images uploaded to its servers for model training. Adobe may consider a paid version in the future depending on user feedback and the success of the experiment.