Pinterest Lens is a camera feature in the Pinterest app that lets users point their phone at any object — a piece of furniture, a clothing item, a food dish — and instantly find visually similar Pins and shoppable products. The system recognizes approximately 2.5 billion objects and processes hundreds of millions of visual searches per month. It was first launched in February 2017 and was significantly upgraded in May 2025 with multimodal search capabilities, allowing users to combine a photo with a text description in a single search.

Pinterest Lens uses deep convolutional neural networks trained on Pinterest's vast image corpus to identify objects, their attributes, and their visual style. The May 2025 update added visual language models (VLMs), enabling the system to process combined text-and-image queries — for example, a user could upload a photo of a jacket and add the text "but in burgundy" to find similar items in a different color. An "animated glow" feature uses AI to generate descriptive words explaining why a user might be interested in a particular Pin, making the visual connection between the search input and the result more transparent. A separate "Shop the Look" feature uses the same computer vision infrastructure to identify and link individual shoppable items within a single lifestyle photograph.