Taylor Kerns / Android Authority
TL;DR
- Google is working to expand Gemini’s markup tools to images generated within Gemini, allowing users to highlight specific areas for editing.
- A pencil icon will soon appear on generated images, opening a markup screen.
- This change will eliminate the need for manual downloads and re-uploads when refining AI-generated visuals.
Marking images to pinpoint generative AI edits is a great quality-of-life improvement for Gemini’s image generation capabilities. We not only spotted the feature in development but also its rollout in December, when Google let users mark up specific parts of the attached image for analysis and editing. But there’s room for improvement: markup tools appear only when you attach an image, not for images Gemini has generated. We’ve now spotted Google working to bring markup tools to Gemini-generated images, too, helping pinpoint further edits.
Within Google app v17.8.59, we managed to enable markup tools on Gemini-generated images.
In the future, when users generate an image in Gemini using Nano Banana, they will see a pencil icon in the top-right corner of the image. Tapping the pencil icon will open the markup screen, where the user can highlight a specific part of the image. Once the user presses Done, the edited image will automatically be attached to the input box, and the user can enter text instructions for further edits.
Don’t want to miss the best from Android Authority?
This greatly simplifies the current approach, as users can continue making changes to generated images without manually downloading them, marking them with an external app, and then re-uploading them (or re-uploading them and marking them in Gemini). It also gives AI specific context about where changes are needed, so the user gets better edits with fewer prompts and without needing to describe the visual elements of the image that need to be edited.
The ability to markup generated images has not yet rolled out in Gemini. We’ll keep you updated when they do.
⚠️ An APK teardown helps predict features that may arrive on a service in the future based on work-in-progress code. However, it is possible that such predicted features may not make it to a public release.
Thank you for being part of our community. Read our Comment Policy before posting.
