Google now lets you try on clothes virtually with just a selfie
Google announced an update to its AI try-on feature on Thursday, enabling users to virtually try on clothes using only a selfie. The company previously required a full-body photo for this function. The enhancement uses Nano Banana, Google’s Gemini 2.5 Flash Image model, to create a full-body digital version from the selfie. This capability launches in the United States today.
The updated feature allows users to select their usual clothing size after uploading a selfie. The system then generates several try-on images based on that size input. Users can review these images and choose one to designate as their default try-on photo. This process simplifies access to virtual fitting without needing extensive photography setup. For those preferring alternatives, the option remains to upload a full-body photo or select from a range of models representing diverse body types. These models provide variety in physique to match different user preferences.
Google initially introduced the AI try-on feature in July. It integrates with the company’s Shopping Graph, which catalogs apparel items available across multiple platforms. Users access the feature by tapping on a product listing or apparel result within Google Search, Google Shopping, or Google Images. Once engaged, selecting the “try it on” icon initiates the virtual fitting experience. This integration spans these core services, making the tool readily available during everyday browsing and shopping activities.
The development reflects Google’s broader investment in virtual AI try-on technologies. The company maintains a dedicated application named Doppl, focused on visualizing outfit appearances through artificial intelligence. Doppl assists users in previewing how various clothing combinations might suit them. Earlier this week, Google enhanced Doppl with a shoppable discovery feed. This feed presents AI-generated videos featuring actual products from merchants. It also includes outfit recommendations tailored to the user’s personalized style profile.
Within the Doppl app’s discovery feed, nearly all content connects directly to purchasing options. Users encounter links to merchants for immediate transactions. The format draws from familiar social media styles, incorporating short videos to display items dynamically. These elements combine to facilitate product discovery alongside virtual try-ons in a seamless interface.