Adobe’s Camera App Leverages Generative AI for a ‘Natural Look’
Adobe’s experimental camera app has taken an intriguing turn. Following the launch of Project Indigo last year, which aimed to deliver a more natural, SLR-like aesthetic for iPhone photography, the Indigo camera app is now being enhanced with a range of generative AI tools. Interestingly, this update does not utilize Adobe’s own Firefly AI models.
The newly introduced “AI Playground” is described as an experimental suite, featuring an opt-out button for users who prefer the app’s original functionality. A select group of Indigo users will have free access to this suite without needing to sign in, as Adobe tests its reception over the coming weeks. Marc Levoy, an Adobe fellow, stated, “Our goal is to learn how people use GenAI-powered editing. Depending on its success, we may extend the experiment, expand the user pool, or run follow-on experiments. If the feature proves popular, we will eventually offer a paid version.”
The AI Playground, included in the version 1.1 update, consists of four main sections: Styles, Object Editing, Photo Guidance, and Custom Edit. Levoy explains that each section features various buttons or toggles that activate GenAI with prompts designed for them. If the outcome of an edit isn’t to your liking, you can simply press the button again to receive a different result.
The Styles section applies preset AI-generated filter effects to photos taken with Indigo, transforming them into illustrations, altering lighting effects, and more. Object Editing provides tools to remove people, objects, and other distractions from images, using generative AI to seamlessly fill in the gaps, akin to Google’s Magic Eraser. Photo Guidance offers feedback by allowing AI to analyze your photos and suggest improvements for reshoots and edits.
The Custom Edit feature is a more flexible tool that enables users to describe specific changes they want to make to an image. For instance, if a Styles preset doesn’t meet your expectations, Custom Edit allows you to detail what adjustments to apply. Initially, Adobe is utilizing Google’s Nano Banana AI model for this update, with plans to potentially incorporate additional models in the future.
This marks a significant shift from the original Indigo experience, which was characterized as a “camera app built by camera nerds for camera nerds.” It leveraged computational photography to help mobile photographers achieve high-quality images that mimic SLR photos, while also offering manual controls for focus, shutter speed, ISO, and white balance. The introduction of generative AI editing features may have been part of the plan from the outset.
“This is the beginning of a journey for Adobe — towards an integrated mobile camera and editing experience that takes advantage of the latest advances in computational photography and AI,” Levoy remarked at the Indigo app’s launch.
Interestingly, the experiment does not reflect the same level of AI transparency that Adobe typically emphasizes. Levoy mentioned that the Indigo team is working on incorporating C2PA Content Credentials — a metadata standard for identifying AI-generated content, already in use by Adobe’s Firefly models — into images edited with Indigo’s generative AI tools. He also noted that Nano Banana applies Google’s invisible SynthID watermark.
“Whether the edits you apply are ‘fair’ or ‘misleading’ depends on what you do, who you share the images with, and how you present them,” Levoy stated. “In the end, nothing can be implemented that reduces the importance of behaving responsibly.”
