Ryan Haines / Android Authority
TL;DR
- Google has upgraded the image editing model in the Gemini app.
- The image editor is now better at maintaining a person’s likeness from one image to the next.
- You can now change a person’s outfit and location, blend multiple photos together to create a new image, make subsequent edits, and apply the style of one image to an object in another.
Back in April, Google rolled out the ability to edit your AI images natively in the Gemini app. Although it’s nice to have that functionality, editing wasn’t exactly perfect, as a subject’s appearance could slightly change from one image to the next. Google is now rolling out an update that promises to fix that problem and introduces some new capabilities.
Don’t want to miss the best from Android Authority?
Today, Google is rolling out a new image editing model to the Gemini app. When developing this upgrade, the tech giant says it was particularly focused on maintaining likeness between subsequent images. So, now when you edit an image, the likeness will stay consistent, regardless of multiple changes.
This update goes hand-in-hand with the new capabilities Google is rolling out for the editor. One advanced editing feature the company has added is changing outfits and locations. When you upload a photo of a person or animal, Google claims the subject will continue to look the same as you put the subject in different clothes and locales.
The next advanced editing feature allows you to blend multiple photos together to create a new image. In the example below, Google shows an image of a person and an image of a dog with the prompt “Create an image where the woman in the photo is cuddling the dog on a basketball court.”
The last new additions are multi-turn editing and design mixing. Multi-turn editing refers to making successive edits to an image. For instance, you can upload an image of a room and ask for the walls to be painted pink. You can then follow that up by adding a bookshelf and furniture without having to start over.
Design mixing, on the other hand, applies the style of one image to an object in another image. In Google’s example, you can see the style of flower petals being applied to boots or the look of butterfly wings being applied to a dress.
After an image is created, you’ll see a watermark in the corner to signal that the image was generated by AI. The image will also be tagged with Google’s invisible SynthID watermark. Google says these watermarks will appear on all images created or edited in the Gemini app.
If you want to give these new features a try, you won’t have to wait long. These advanced editing features are rolling out starting today.
Thank you for being part of our community. Read our Comment Policy before posting.
