Google is updating its AI try-on characteristic to allow you to nearly attempt on garments utilizing only a selfie, the corporate introduced on Thursday. Up to now, customers needed to add a full-body image of themselves to nearly attempt on a chunk of clothes. Now they will use a selfie and Nano Banana, Google’s Gemini 2.5 Flash Picture mannequin, to generate a full-body digital model of themselves for digital try-ons.
Customers can choose their normal clothes dimension, and the characteristic will then generate a number of photos. From there, customers can select one to make it their default try-on picture.
If desired, customers nonetheless have the choice to make use of a full-body picture or choose from a spread of fashions with various physique sorts.
The brand new functionality is launching in america immediately.
Google first launched the try-on characteristic in July, permitting customers to attempt on attire gadgets from its Procuring Graph throughout Search, Google Procuring, and Google Pictures. To make use of the characteristic, customers must faucet on a product itemizing or attire product consequence and choose the “attempt it on” icon.
The transfer comes as Google has been investing within the digital AI try-on area, as the corporate has a separate app devoted particularly to that function. The app, known as Doppl, is designed to assist visualize how completely different outfits may look on you utilizing AI.
Earlier this week, the tech large up to date it with a shoppable discovery feed that shows suggestions so customers can uncover and nearly attempt on new gadgets. Almost every little thing within the feed is shoppable, with direct hyperlinks to retailers.
The invention feed options AI-generated movies of actual merchandise and suggests outfits primarily based in your customized fashion. Whereas some might not be keen on an AI-generated feed, Google doubtless views it as a approach to showcase merchandise in a format that persons are already conversant in, due to platforms like TikTok and Instagram.
