Google is developing an annotation feature for Gemina images

Google is developing a new feature for Gemina that will allow users to “write margins” on images.

This feature allows users to draw or add text to an image before downloading it, and possibly resubmit annotated images to have their desired changes applied to it. According to reports, the purpose of this feature is to increase user control and facilitate direct interaction with AI results.

Ability to annotate Gemina’s images

According to the information published by TestingCatalog on the X social network, Gemina is developing a new feature that allows advanced annotations beyond simple text writing for users. With this feature, users can add drawings and text on the produced images. The leaked image of the feature shows a horizontal color palette with two main tools at the bottom: a squiggle for drawing and a ‘T’ for inserting text.

Another user also reported that it is possible to repost images after annotating them. He explains that users first download the image, put their drawing or text instructions on it, and then apply the desired changes. This method allows users to make minor changes directly to the AI ​​output without having to re-enter the original prompt.

Since the beginning of this year, Google has launched the ability to edit images internally in Gemina. This tool can be used not only on images produced by artificial intelligence, but also on users’ photos, and offers features such as changing the background, adding or removing objects, and combining multiple images. Also, users can tell a story and artificial intelligence tries to create harmonious images for it.

Recently, Gemina’s image capabilities have been expanded with the introduction of the Nano Banana Pro model. Google has announced that this model is able to produce more detailed images with richer content and improve the clarity of text and fonts.