According to healthdata.org unhealthy eating accounts for approximately 680'000 deaths in the US alone and is a major risk factor for heart diseases, diabetes and high blood pressure. But what exactly is unhealthy eating? The three major unhealthy eating habits include consuming highly processed food such as fast food, too much sugar and sodium and having little to no diversity in food intake. Most of the people have no idea how many calories they are eating in a whole day, let alone do not even know how many calories would be enough to maintain a healthy life. Second, for most it is too tedious and time consuming to come up with healthy recipes. FoodAIe was born to tackle these problems, helping our fellows to eat health in a simple, fun and entertaining way.
What it does
Our app has two main functionalities: First, making use of artificial intelligence and computer vision algorithms, FoodAIe let's you track your calorie intake and nutritious value of your meals by simply taking a picture of your food. Second, FoodAIe makes use of its extensive knowledge, extracted from several APIs and Databases, to give you recipe recommendations based on the food you have at home, what you already ate on this day and based on food restrictions, special diets and calorie goal of the day. In order to increase user experience and engagement, a networking platform is integrated where you can track how your friends are doing with their diet and you can unlock various achievements based on how healthy your eating is.
How we built it
The app was built using React-Native for the frontend framework and a Flask API for the backend. To provide the users with large amounts of information and of AI algorithms, we have profited from the food recognition API generously provided by Bite.ai (https://bite.ai/). Furthermore, we used Bite.ai and Migros APIs for Product and Recipe recommendations.
Challenges we ran into
No one of our team had great experience with using React-Native for mobile app development which meant that it took us a lot of time to get the frontend done. Furthermore, being a hybrid team (combine online and offline participants) posed a challenging, though exciting experience.
Accomplishments that we are proud of
We are proud that we managed to develop a first working prototype of the app in such a short amount of time.
What we learned
We learned coordinating and working in a group that is physically not in the same place. What's more we learned how to build apps with React-Native which was not the best idea to do during a hackathon but it worked.
What's next for Camera-based nutrition and diet app
Despite a great progress, the scope of FoodAIe can be easily extended. The AI algorithms could be further refined in order to provide the user with even larger amounts of information about their food products and their nutrients.