Valentyn Volkov / Shutterstock

MIT’s New App Uses Artificial Intelligence to Recommend Recipes Based on Food Pictures

The paper on the AI system will be presented at the Computer Vision and Pattern Recognition conference

Valentyn Volkov / Shutterstock

The system uses a database of food pictures and recipes to make recommendations.

Instagram has grown to be the perfect platform to share your pictures of delicious meals, so much that one restaurant in London even capitalized on the foodie trend with its very own Instagram kits. Aside from food bloggers and influencers who might share how to make the dish, most of us give food pictures a casual “like” and then it’s on to the next. But now, thanks to Massachusetts Institute of Technology, there’s an app that can analyze photos and let you know how to make the meals you see on your Instagram feed.

Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory and the Qatar Computing Research Institute created Pic2Recipe, an app that uses artificial intelligence to predict ingredients and suggest similar recipes based on looking at a picture of food.

According to MIT, the app does well with desserts, but has difficulty with foods such as sushi rolls and smoothies, and with distinguishing between similar recipes for the same dishes.

In the future, researchers hope to improve the system to understand images of food in more detail, including identifying cooking and preparation methods. They are also interested in recommending recipes based on dietary preferences and available ingredients.

In addition to providing recipe recommendations, the app can also provide deeper insight on people’s eating habits.

“This could potentially help people figure out what’s in their food when they don’t have explicit nutritional information,” Nick Hynes, graduate student and lead author of the paper about the artificial intelligence system.


“For example, if you know what ingredients went into a dish but not the amount, you can take a photo, enter the ingredients, and run the model to find a similar recipe with known quantities, and then use that information to approximate your own meal.”