You see it nearly every time you go out to eat.
Someone whips out their phone and snaps a picture of their meal.
Now foodies can put those pics to good use instead of just using them to collect "likes" on social media.
A new app can turn those food pics into recipes!
Behold Pic2Recipe from the brainiacs at MIT.
Researchers at the school's Computer Science and Artificial Intelligence Laboratory created a neural network to find connections between more than 1 million images and recipes.
Their demo vid shows how the app correctly identified the recipe for sugar cookies, except it missed three ingredients.
The Verge put the demo version through a short round of testing and reports it couldn't identify potato chips, bowls of ramen or rice and beans.
Delish points out the app isn't ready for the market just yet.
MIT researchers add that they're going to keep working on it so that the app can predict how the dish should be prepared. For example, diced versus chopped.
They also want the app to learn variations such as beefsteak tomato versus cherry tomato.
Researchers add it'll be a good way for people to track daily nutrition by seeing what's exactly in their food.