Google Upgrades AI Try-On: Snap a Selfie, Try Any Outfit
Now use a selfie, not a full-body shot, to instantly visualize outfits.
Google is removing the biggest friction point in virtual shopping: the need for a full-body photo. Its updated AI try-on feature now turns a simple selfie into a personalized, full-body model you can dress in any apparel listing across Search, Google Shopping, and Google Images. Powered by the Gemini 2.5 Flash Image model, internally known as Nano Banana, the system generates multiple photorealistic try-on images based on your selected size, letting you pick the best result to save as your default.
This upgrade builds on the try-on tool Google launched in July, which lets shoppers tap “try it on” on apparel product listings to see how items look. The selfie-first approach dramatically lowers the barrier to entry—no mirrors, no lighting setups, just your camera and a quick snap. For users who prefer it, the original full-body upload option and a diverse set of model presets remain available, so you can still compare across different body types.
The timing is significant. Just this week, Google updated its dedicated AI try-on app, Doppl, with a shoppable discovery feed that surfaces personalized outfit recommendations. The feed blends AI-generated videos of real products with direct merchant links, resembling the short-form, shoppable content people scroll daily on TikTok and Instagram. In practical terms, this means you can discover, visualize, and purchase up to speed, from inspiration to checkout, with far fewer steps.
For shoppers, there are three immediate wins. First, accuracy improves because your virtual model now starts from your actual proportions and posture via selfie capture. Second, decision velocity increases: multiple generated images per try-on let you evaluate fit and style quickly. And third, accessibility expands: you can use the feature anywhere in Google’s shopping surfaces, with the Doppl app offering a deeper, feed-driven discovery experience.
For merchants, this is a pivotal moment to optimize. Ensure your product imagery is high-resolution, properly lit, and includes detailed views—Google’s AI modeling relies on clean inputs. Maintain accurate size data and fit notes in your product feeds; mismatches between listed sizing and visual outcomes can erode trust. Encourage UGC photos and honest reviews to complement AI visualizations, reinforcing authenticity and conversions.
There’s also an E-E-A-T angle for brands. As AI-powered visuals become standard, trust hinges on transparency. Clearly communicate that images are AI-generated and base sizing on verified measurements. The brands that educate rather than embellish will see better engagement and fewer returns.
As of today, the updated selfie try-on is rolling out in the United States. Whether you’re testing everyday basics or statement pieces, the path from discovery to confident purchase just got shorter. Add a clear photo, pick your size, tap “try it on,” and let the mirror you do the shopping.
No Comments