LettuceEat

LettuceEat Architecture
LettuceEat Log-in Screen
Main Challenge Page
Main Challenge Page 2
Challenge a Friend Feature
Achievements Page
Rewards for Social Good
Redeem Details
Redeem User Prompt
Redeem Confirmation Page
CDC

Inspiration

Humans are creatures of habit. We often end up walking the same well-worn paths without thinking about it, and food is no different. We’re interested in building an app that shakes people from their routine diets, is fun to use, and helps them make healthier food choices - one day at a time. We also found research by the Center for Disease Control and Prevention that indicated that a large amount of Americans are not incorporating fruits in their daily diets. While there are lots of tools for calorie counting and serious dieting, we found a gap in low-pressure, fun-to-use apps that helped people explore new foods and eat better in attainable ways. Moreover, we also wanted to broaden the impact of each day’s decision to the rest of the world by incorporating rewards mechanisms that allow users to donate to charity.

What it does

LettuceEat helps users take on one food challenge a day. Users can accept a challenge and get small reminders throughout the day, usually at mealtimes, to complete their challenge. To complete each challenge, users will upload a photo of their food item to the app, which will then scan the photo using Microsoft Custom Vision Service API to check that the food item fulfills the challenge requirement. Users can also see their friends’ activities in a feed, or compete against a friend for a given challenge. Upon completing a challenge, users will receive points that they can use towards redeeming philanthropic rewards, such as a donation to a food-related foundation.

How we built it

We began by creating a storyboard that detailed a typical user’s journey through the app. We found that our main goals were to make it easy and fun for people to complete food challenges, and to promote social good through individual decision-making. We moved on to simple paper sketches that helped us flesh out the details of our app quickly. We wanted to incorporate a cute lettuce icon designed by Mis0Happy in our prototype as an example of a potential mascot as a way to create empathy with the user and to encourage them to try new things. Then we began building and wireframing using Google’s material design tools and principles. We built the site on Microsoft Azure and utilized Microsoft’s Custom Vision Service API to provide image recognition and analysis of user photos. We also trained the Microsoft’s Custom Vision Service API to get better at recognizing fruit and vegetables rather than other aspects of the image.

Challenges we ran into

This was our entire team’s first hackathon. We ran into a ton of issues, large and small, but we learned so much. One challenge was to reduce the large amount of information that Microsoft Computer Vision API returned for each image and simplify the information so that it was usable for our app. We also spent a lot of time figuring out how to capture real-time pictures from mobile; coding Custom Vision API to recognize the photo and giving users visual feedback while analyzing their data quickly on the cloud and rendering in real-time was difficult. We enabled Azure CDN on Azure Blob Storage to accelerate the loading time. Lastly, allotting time appropriately for developing the most important features was another major challenge.

Accomplishments that we're proud of

Getting set up on Azure. Being able to upload to the cloud and download into the stream. Building a working prototype with Computer Vision Services API. Our user interface. Designing a tool that helps individuals make small steps to better themselves and better the world.