Facebook is sharing details on how the company improved AI and expanded GrokNet, a product recognition system powering shopping features on Facebook, like product tagging and showing visually similar products based on the seller’s catalogue and the upcoming visual search to Instagram.
Facebook is also driving its AI advancements that provide an understanding of product attributes and multimodal signals, to further develop shopping on the platform. Facebook AI is enhancing the social media platform with the visual search capabilities of GrokNet. GrokNet identifies what products are in an image and predicts their categories, like sofa, and attributes, like color and style.
Unlike previous systems, which required separate models for each vertical, GrokNet is an all-in-one model that scales across billions of photos across different verticals, including fashion, auto, and home decor. GrokNet started as a fundamental AI research project with its first few applications on Marketplace, where AI analyzes search queries like “midcentury modern sofa” and predicts matches to search indexes so that the users who visit Marketplace get the most relevant results when searching for products.
Since 2020, Facebook has expanded this technology to new applications to make posts more shoppable across new Facebook applications. Right now, when a seller posts an image on their Facebook page, the AI-powered shopping system helps identify untagged items and suggests tags based on their product catalog — so instead of manually tagging their items, a seller could create and post their photo in just seconds. And when a shopper is viewing an untagged post from a seller, the system suggests similar products below the post from that seller’s product catalog.
Facebook also shares details on the newest advancements that are making AI systems better at recognizing products — from multimodal understanding to learning deeper, more nuanced attributes.
For example on Instagram, shopping begins with visual discovery. Users scroll through the app and see a floral dress for summer or a wedding dress, and with AI-powered visual search, people could find similar dresses by tapping on an image they see within Instagram.