Get all your news in one place.
100’s of premium titles.
One app.
Start reading
Geekflare
Geekflare
Keval Vachharajani

Gemini Got a Big Image Editing Upgrade: Here’s How to Use It

Google has announced a big update to its image editing model, developed by DeepMind. It now promises more consistent results, particularly when editing people or pets. The upgrade is designed to fix one of the biggest frustrations in AI image editing: characters looking “almost right” but not quite the same.

With the new model, Gemini can now preserve likenesses across multiple edits. Which means it now allows you to put your friend in a Halloween costume, see your dog in a tutu, or preview yourself with a retro hairstyle, without losing key details that make the subject recognizable.

What’s New in Gemini’s Image Editing? 

The new update brings some interesting capabilities to image editing. The app can now maintain consistent character design, ensuring that people or pets look the same across different edits. It also enables creative composition and allows you to merge images, styles, and objects into a single cohesive picture. The multi-turn editing makes it possible to work, for example, repainting a room, adding furniture, and later swapping the wallpaper without starting over. 

Furthermore, style adaptation allows you to apply patterns, textures, or colours from one image to another, while precise local edits allow for small but important adjustments such as changing a tie’s colour or removing background objects. However, the interesting part is that once edited, images can even be re-uploaded to Gemini for video generation. Besides, every creation comes with both a visible watermark and DeepMind’s invisible SynthID digital watermark, marking them as AI-generated.

Google’s Tips for Better Results 

Google has also shared best practices for crafting prompts in Gemini to get the best results. The company recommends being clear about six elements: subject, composition, action, location, style, and editing instructions. Adding these details helps Gemini produce more accurate and creative results.

For example, instead of just saying “a cat with a hat”, Google suggests specifying: “a fluffy calico cat wearing a wizard hat, wide shot, standing in a sunlit meadow, in a watercolor style.”

When it comes to the availability part, you can try this updated image editing capability in the Gemini app starting today.

Sign up to read this article
Read news from 100’s of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.