Google has introduced a revolutionary update to its AI assistant, Gemini, giving users unprecedented control over their photos. Developed by Google DeepMind and touted as the world’s highest-rated image editing model, the new feature is available to all users today through the Gemini app. The update not only preserves faces but also offers professional tools such as photo merging, step-by-step editing, and transferring styles.

Gemini Solves a Key Problem: Faces Will No Longer Be Distorted

One of the most common problems in AI image editing was that people in edited photos didn’t exactly resemble the original. Google focused on solving this problem of close but not quite. The new model consistently preserves the character’s original facial features and appearance when placing a photo of a person or pet in a different scenario. This means the results are surprisingly realistic, even when you imagine yourself in a different costume or in a different time period.

The most exciting part of the update is the powerful new editing capabilities offered to users. Here are some of the highlights:

Photos Combination: You can now upload multiple photos to create a completely new scene. For example, combine a photo of yourself with a photo of your dog and show the two of you on a basketball court.

Step-by-Step Editing: Gemini now remembers the editing process. You can start with a photo of an empty room, paint the walls, add a bookcase, and finally place the furniture. Gemini works with you, preserving previous changes at each step.

Design and Style Transfer: This feature allows you to apply the texture or pattern of one object to another. For example, you can design a dress using the pattern on a butterfly’s wings or transfer the texture of a flower to a pair of rain boots.

Google also prioritizes transparency in AI-generated content. All images created or edited in the Gemini app will include both a publicly visible watermark and the invisible SynthID digital watermark. This ensures it’s always clear that the images were generated by AI. These powerful and creative features are available to try in the Gemini app starting today.