this january, they upped their game, though, launching pinterest lens, a way of being able to searcha attached. and from that search term, it aims to come up with similar objects and related ideas. there we go, we've got a picture and something is emerging. it says "shoes." right, those are definitely shoes, but they don't exactly resemble mine. black shoe with blue laces, some men's shoes. so how does it actually do this? so there are two parts to visual search. the first is computer vision, which is a way of translating the information coming in through the camera into words. the second is the data set and the data set is the most important. so with pinterest lens, when you point your camera at something in the real world, the computer inside the phone translates that image into text and then it takes the text and it takes the image and runs a search against 100 billion pins on pinterest to find the ones that seem the most relevant. 0k, it knows it's a lemon, that's definitely a good start. now, if we scroll down. ok, i think you scroll through it and some of the results make sense