Facebook’s new AI tool will automatically identify items you put up for sale

Author: Web Desk

Facebook is launching what it’s calling a “universal product recognition model” that uses artificial intelligence to identify consumer goods, from furniture to fast fashion to fast cars.

It’s step one towards a future the place the merchandise in each picture on its web site might be recognized and doubtlessly shopped for. “We want to make anything and everything on the platform shoppable, whenever the experience feels right,” Manohar Paluri, head of Applied Computer Vision at Facebook, instructed The Verge. “It’s a grand vision.”

This can be quite challenging, since it can get difficult to optimize parameters for one task without reducing the effectiveness of another. For instance, optimizing a model to recognize cars well might mean that it can’t recognize patterns on clothing as effectively. “I want to build something that ultimately feels like having one of your best friends with you whenever you get dressed, shop, want a recommendation, or simply need some inspiration,” Tamara Berg, a Facebook research scientist, wrote in a blog post. Facebook says it essentially wants to use these AI techniques to transform online shopping for everyone on its platforms through personalization. It’s currently unknown when we’ll start to see these models come into play on the platform, but it’ll certainly be helpful once they do.

Facebook says what makes its instruments totally different are their scope and accuracy. The firm’s new product recognition tool, GrokNet, can identify tens of hundreds of various attributes in a picture. These vary from particular manufacturers to issues like colour and dimension.

GrokNet has already been deployed on Facebook Marketplace, the place it helps customers shortly checklist items for sale by figuring out what’s in them and producing quick descriptions. You would possibly add a photograph of your sofa, for instance, and Marketplace will recommend itemizing it as “black, leather, sectional sofa.”

The firm can be testing a model of this tool that’s constructed for companies. When they add images to their web page containing their very own merchandise, the AI system can automatically tag them and hyperlink to purchasing pages.

It’s unclear, though, exactly how accurate GrokNet is. The company says it can identify 90 percent of images on Marketplace in the Home and Garden category, but it didn’t give similar statistics for other types of product categories.

As is often the case with tools like this, the difference between the advertised features and actual user experience can be huge, and we’ll have to wait and see what reaction GrokNet gets from Facebook’s users.