Shopping for clothes can be a frustrating experience for many people—especially when it comes to trying on garments. While it’s entirely logical to test how a piece fits before buying, not everyone enjoys the process. Many shoppers would prefer a faster, more seamless experience without the need to step into a fitting room. Google Labs has taken a significant step toward making that possible with the launch of Doppl, an experimental new app.
What is Doppl and How Does It Work?
Doppl is Google’s new virtual fitting room technology, first unveiled during the company’s Google I/O 2025 developer conference. It uses artificial intelligence to let users visualize how clothing would look on their body—without ever putting it on physically.
The app allows users to create a personalized avatar using a full-body photo. Then, through AI and motion simulation, Doppl animates how different clothing items behave and fit on the avatar. Whether it’s a dress, t-shirt, or pair of jeans, Doppl gives a near-realistic preview by animating the garment’s movement and fit.
From Social Media to Smart Dressing
Doppl isn’t limited to items found on shopping websites. It can also use clothing images from Instagram, online stores, or even screenshots. This versatility allows users to preview how almost any outfit might look and move on their own body type, bridging the gap between inspiration and decision-making.
Each interaction feeds the app’s learning algorithms, helping refine its predictions and improve garment realism. The more people use it, the better it gets at simulating complex fabrics, unique cuts, and various body shapes.
Still Experimental, But Promising
Despite its innovation, Doppl is still in development and has a few limitations. For example, garments with intricate textures or unusual silhouettes might not be rendered perfectly. Nonetheless, this is expected at the early stage of such advanced tech.
Currently, Doppl is only available in the United States for both iOS and Android users. Its broader release will depend on user feedback and further testing.