Key Takeaways

  • Adobe’s Project Indigo app adds AI features for photo critiques.
  • New tools include advanced object removal and depth of field generation.
  • Users receive suggestions for framing and editing their photos.
  • Features are currently in testing and may not be widely available.

New AI Features in Project Indigo

Adobe is enhancing its experimental iOS camera app, Project Indigo, with new AI-driven capabilities. Launched last year, the app previously included professional controls and multi-frame super-resolution. The latest update introduces features that utilize large language models to critique photos and suggest edits.

Photo Critique and Editing Suggestions

The app now offers a photo critique feature that provides a “professional” assessment of various aspects such as framing, lighting, colors, and emotional impact. Additionally, users receive capture and editing suggestions, guiding them on how to improve their shots. For example, the app advised removing an unwanted object from a photo I took.

Advanced Object Removal

Project Indigo also enhances object removal capabilities. Unlike traditional methods that require users to manually select objects, this app allows for toggling specific items to be removed, such as background people, trash, or vehicles. The results are impressive, as the app successfully removed a friend and an object he was holding without leaving artifacts.

Depth of Field and Style Transfer

Another feature allows users to create a simulated depth of field, blurring backgrounds for a more professional look. Adobe is also experimenting with style transfer, enabling users to apply artistic effects reminiscent of watercolor or monochromatic styles. While the outputs are refined, style transfer itself is not a groundbreaking feature.

Testing Phase and Future Prospects

Adobe is currently testing these features using Google’s Gemini-based Nano Banana model, with the option to incorporate its own Adobe Firefly model. These AI tools are still in the experimental stage, and access is limited to select users. While it remains uncertain if these features will be rolled out to a broader audience, they aim to enhance users’ understanding of photography nuances rather than simply generating standard edits.