The CSAIL team has created a database of over 1 million recipes. They used that data to train a neural network to find patterns and make connections between the food images and the corresponding ingredients and recipes. The project aims to build off of this work but dramatically expand in scope. (Pic2Recipe)
The system did particularly well with desserts like cookies. It had difficulty determining ingredients for more ambiguous foods, like sushi rolls and smoothies. It was also often stumped when there were similar recipes for the same dishes. The team has an online demo where people can upload their own food photos to test it out.
The could infer how a food is prepared. The researchers are also interested in developing the system. The project was funded, in part, by QCRI, as well as the European Regional Development Fund (ERDF) and the Spanish Ministry of Economy, Industry, and Competitiveness.
Fastfood5
Artificial intelligence suggests recipes based on food photos