May 5, 2025
Most online searches start with words, but describing a vibe, style or aesthetic can be hard and limiting. As a visual search platform, Pinterest enables people to overcome this obstacle, by using images at the starting point for a search.
Today, Pinterest is introducing a set of new visual search features that allows users to further interact with image Pins so they can more intuitively discover, refine and shop the ideas that perfectly match their personal tastes. These tools will be available for women’s fashion content across the U.S., Canada and the U.K., and will expand to more categories and countries over time.
“Our visual search technology represents a shift in how users interact with and discover inspiration,” said Dana Cho, Pinterest VP of Design. “We're not simply delivering search results—we're curating a personalized journey of discovery that empowers individuals to find their unique style, and shop it too."
We’re breaking down and decoding images so users can quickly search and shop for the details of an outfit. Whether it’s an overall aesthetic, a color palette, a specific fit, or product category, when users view a Pin, we’ll now generate the words they can use to figure out what they like about the image, and then further explore and shop. To help users more easily identify and select the objects that they want to see and shop more of, we’ve added a new animated glow.
Ever come across something you love but want it in a different color, style, or fabric? We’re adding a new refinement bar that’ll help users narrow their search results, so they can find exactly what fits their personal style. For example, you spot an outfit you like, and use the “style” refinement to see similar outfits but more y2k, or you love this blue blazer, and use the “occasion” refinement to see more formal options.
We’re expanding where users can access our visual search features, including the “Search image” button, by enabling users to simply long press on any Pin on their home feed.
Pinterest is able to deliver this rich search experience by leveraging our multimodal visual refinement technology at scale. We use Visual Language Models (VLMs), a form of generative AI, to provide users with a more expansive visual "vocabulary" to "describe" their style ideas. And we've integrated industry-leading multimodal embedding models to provide users with the tools to more easily search using a combination of image and text to find the exact content matching their specific tastes.
No more settling for generic search results or feeling limited by not having the right words. With our new visual search features, users can more intuitively discover, refine and shop, turning their inspiration into reality.