Google expands AI clothing try‑on to work with selfies

Key Points

  • Google’s AI try‑on now works with a single selfie.
  • Uses the Nano Banana Gemini 2.5 Flash Image model to create a full‑body avatar.
  • Users select size, view multiple generated outfit images, and set a default try‑on photo.
  • Option to upload a full‑body photo or choose from diverse model representations remains.
  • Feature integrated across Google Search, Shopping and Images.
  • Doppl app receives a shoppable discovery feed with AI‑generated videos and style suggestions.
  • Launch limited to the United States.
  • Goal is to make online fashion shopping more interactive and personalized.

Google’s AI try-on feature for clothes now works with just a selfie
Content image from Google’s AI try-on feature for clothes now works with just a selfie

New selfie‑based try‑on capability

Google revealed that its virtual clothing try‑on feature now allows users to start with just a selfie. The system leverages Google’s Nano Banana model, part of the Gemini 2.5 Flash Image family, to generate a full‑body digital version of the user. This avatar can then be dressed in apparel items drawn from Google’s Shopping Graph, letting shoppers see how garments might look on them without needing a full‑body photo.

How the feature works

After taking a selfie, users select their usual clothing size. The AI creates several images of the selected outfit on the generated avatar. Shoppers can review these options and set one as their default try‑on photo. For those who prefer a traditional approach, the option to upload a full‑body picture remains available, as does the ability to choose from a library of models representing a range of body types.

Integration with Google services

The updated try‑on experience is integrated across Google Search, Google Shopping and Google Images. When a user taps the “try it on” icon on a product listing, the AI-powered overlay appears, offering an immediate visual of the clothing on the user’s avatar. This follows Google’s initial launch of the feature, which first appeared in July and enabled virtual try‑ons for apparel items across the same platforms.

The Doppl app and broader AI fashion push

Google also operates a dedicated app called Doppl, designed specifically for AI‑driven fashion visualization. Recently, Doppl received a shoppable discovery feed that showcases AI‑generated videos of real products and suggests outfits based on personal style. Nearly every item in the feed links directly to merchants, reinforcing Google’s strategy to blend AI visualizations with seamless purchasing pathways.

Implications for shoppers and retailers

By allowing a selfie‑based try‑on, Google lowers the barrier for consumers to experiment with online fashion, potentially increasing conversion rates for retailers. The technology also reflects a broader industry trend toward immersive, AI‑enhanced shopping experiences that mimic familiar social media formats.

Availability

The new capability is launching in the United States today, extending Google’s suite of AI tools that aim to make online apparel shopping more interactive and personalized.

Source: techcrunch.com