Indian food is one of the most popular cuisines in the world and comprises uniquely flavored preparations using traditional recipes and spices. The large number of states, billion plus population, and thousands of years of culinary heritage of India has resulted in large number of sub-cuisines and vast number of unique dishes.

To aid Indian food enthusiasts, such as myself, in their quest for wellness and fitness, I have conceptualized CurryAI – a computer vision aided Indian food nutrition calculator. This calculator would be able to decipher the nutritional content of an Indian dish by means of analyzing an image of the dish.

The artificial intelligence within the calculator could be embedded as part of a mobile app, easily enabling a user to get a dish’s nutritional content after snapping a picture from their phone.

In researching literature for a similar capability, I found prior research performed at Google titled “Im2Calories”. This paper approaches the problem of detecting calories from images, not complete nutritional information. It also uses information from menus of popular US restaurants, thus reducing the problem to classification of a dish within the context of a known restaurant’s menu. This makes the problem far more simplified and tractable.

My research focuses on the problem of generic meal detection as that is more widely applicable.

I could not find any dataset, paper or commercially available apps that focused on identification of Indian food. I have attempted to address this gap by mean of my research.

Thus, I used nutritional breakdown of a dish by using information from the Food Data Central US government website (https://fdc.nal.usda.gov/). The Food Central Database contains information about a large variety of dishes, including many Indian dishes.

I also created a mapping of Indian food dishes to their nutritional content with respect to ingredients, calories, macro nutrients (protein, carbohydrate, fats, fibre etc.), using Food Data Central database.

CurryAI highlights important steps in providing complete nutritional information for Indian dishes.

There are many other pieces to this problem, such as plate segmentation and portion size detection, which can be addressed through future work. This work can also be extended beyond Indian food to global food recognition.