Google Shopping has launched a new virtual try-on experience for online shoppers, which utilizes generative AI

Google Shopping has recently unveiled a range of exciting features aimed at enhancing the online shopping experience. One of the most notable additions is a virtual try-on experience that allows users in the US to visualize how clothing will look on different body types. This innovative feature includes real human models representing diverse skin tones, ethnicities, hair types, and body shapes, ensuring a more inclusive representation.

Initially, the virtual try-on experience is limited to women's tops from select brands such as H&M, Anthropologie, Everlane, and Loft. However, Google plans to expand this feature to include men's tops and other apparel later this year. The primary objective of the virtual try-on feature is to assist shoppers in avoiding disappointment by providing an accurate visual representation of clothing before making a purchase. Google cites data indicating that a significant percentage of online shoppers express dissatisfaction with their clothing purchases and feel underrepresented by online models.

To achieve a realistic depiction of clothing, the virtual try-on experience utilizes a diffusion-based generative AI model. This advanced AI model learns to remove random pixels, known as Gaussian noise, added to an image, resulting in a more lifelike representation of how the clothing will look on different models. It's important to note that the AI shapes the clothing around images of human models, and the models themselves are not AI-generated.

In addition to the virtual try-on feature, Google Shopping has introduced new filters to help users find specific items within its platform. These filters enable users to search for cheaper alternatives, different patterns, or specific styles. Machine learning and visual matching algorithms power the filtering feature, allowing users to refine their search based on color, style, and patterns across multiple online clothing stores. Currently, the filtering feature is limited to tops and can be accessed within the product listings on Google Shopping. However, there is no mention of when this feature will expand to include other types of apparel. As the platform continues to evolve, it will be interesting to see how these features expand to encompass a broader range of products and further change the way we shop online.