Inspiration

Currently, there is an exponential growth of obesity in the world, leading to devastating consequences such as an increased rate of diabetes and heart diseases. All three of our team members are extremely passionate about nutrition issues and wish to educate others and promote healthy active living.

What it does

This iOS app allows users to take pictures of meals that they eat and understand their daily nutrition intake. For each food that is imaged, the amount of calories, carbohydrates, fats and proteins are shown, contributing to the daily percentage on the nutrition tab. In the exercise tab, the users are able to see how much physical activities they need to do to burn off their calories, accounting for their age and weight differences. The data that is collected easily syncs with the iPhone built-in health app.

How we built it

We built the iOS app in Swift programming language in Xcode. For the computer vision of the machine learning component, we used CoreML, and more specifically its Resnet 50 Model. We also implemented API calls to Edamam to receive nutrition details on each food item.

Challenges we ran into

Two of our three team members have never used Swift before - it is definitely a challenge writing in an unfamiliar coding language. It was also challenging calling different APIs and integrating them back in Xcode, as the CoreML documentation is unclear.

Accomplishments that we're proud of

We are proud of learning an entirely new programming language and building a substantial amount of a well-functioning app within 36 hours.

What's next for NutriFitness

Building our own machine learning model and getting more accurate image descriptions.

Built With

Share this project:

Updates