Pic2Recipe Uses a Neural Network to Predict Recipes Based on Photos

'Pic2Recipe' is the name of a new recipe tool developed by MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) to supply a list of ingredients for food photos based on image recognition.

Since it is quite common for a person to come across a delicious-looking dish in a photo that is not accompanied by a recipe, Pic2Recipe helps people analyze images. Amazingly, the tool is able to boast a 65% success rate when it comes to predicting the individual ingredients in a recipe.

Some of the most popular recipe additions include staple ingredients like butter, sugar, olive oil, water and eggs, with most recipes featuring an average of nine ingredients. The potential uses of Pic2Recipe are limitless, meaning that in the future, people may be able to use the tool to track their nutrition goals, make dishes at home or identify recipes that have ingredients that conflict with their diets or food preferences.