Google is enhancing its use of generative AI through additions to its Shopping and Search features designed to help shoppers find items that fit their needs. The new options range from virtual try-ons to curated review snapshots to make it easier to find the perfect product.
Google’s shopping capabilities will be enhanced with the edition of a virtual try-on tool that uses generative AI to show shoppers what a top would look like on a diverse set of real models with an emphasis on accurately reflecting subtle but crucial fabric details.
“Perhaps the most popular reference for virtual try-on dates back to the movie Clueless,” said Ira Kemelmacher-Shlizerman, Senior Staff Research Scientist for Shopping at Google in a blog post explaining the new tech. “We’ve come a long way since Cher’s closet, though. Current techniques like geometric warping can cut and paste and then deform a clothing image to fit a silhouette. Even so, the final images never quite hit the mark: clothes don’t realistically adapt to the body and they have visual defects like misplaced folds that make garments look misshapen and unnatural. So when we set out to build a new virtual try-on feature, we were committed to generating every pixel of a garment from scratch to produce high-quality, realistic images.”
The virtual try-on technology will initially be available for women’s tops from brands including Everlane, H&M, Loft and Anthropologie, with the list of partners and styles set to expand in the future.
The new Shopping feature also has added “guided refinements,” which seek to emulate online the in-store experience of associates finding complementary accessories. This tool works as a filter that lets shoppers fine-tune their choices until they find the piece that fits their needs.
Shoppers also can find the right retail stores while traveling with the help of a new Search feature that lets them ask detailed questions about their destination to receive AI-powered “snapshots” with information from across the web, reviews and photos on Google.
The enhanced Search features also hold promise for shoppers looking to purchase something closer to home. For example, a search for Bluetooth speakers could produce an AI-powered snapshot that includes insights and products that can help shoppers find exactly what they need at the right price point. This generative AI Search experiment will initially be available to those who sign up through Search Labs.
Both the Shopping and Search features work with Google’s Shopping Graph, which has more than 35 billion product listings that it uses to power a dataset of constantly changing products, sellers, brands, reviews and inventory. Every hour, more than 1.8 billion listings are refreshed in Shopping Graph to help enable reliable results.