Pinterest has added image-recognition software to its site which will allow users to search online for the contents of the site's images.
This is done via deep learning technology, and users can draw a box around an item present in an image to search for similar items among an index of more than one billion. Some will also be available to purchase with Pinterest's recently-introduced Buy button.
Speaking to MIT Technology Review, Kevin Jing, engineering manager for Pinterest, said: "Image representation coming from deep learning is much, much more accurate. Even this year there has been so much improvement."
Earlier this month, Facebook unveiled its own deep learning project, which similarly uses artificial intelligence to achieve advanced object recognition that is able to segment and distinguish between items in an image. With the announcement, the company claimed that the software is 30 per cent faster than previous attempts.
"Sometimes you spot something you really love on Pinterest, but you don't know how to find it in real life, or what it's even called," Jing wrote in an official blog post. "There's that perfect lamp hiding in a Pin of someone's living room, or maybe a random street style with the exact shoes you're looking for.
"Well, now we've got a new tool that lets you find all those things you don't have the words to describe."
