• 

  Our skincare app's logo; reflects the core purpose of our application

• 

  A snapshot of our Roboflow model training process

🧠🔥 What sparked this idea in our minds?

At the many hackathons that we've been to (collectively), we have seen incredible projects that solved real-world problems; however, a problem that we found nearly untouched and unheard of at these events was the concept of good skincare of the process(es) to achieve it! Thus, scancare was born.

🤳 What it does

scancare provides simple but vital habits and routined action items that users can add onto themselves. Our app features a camera which users utilize to capture a photo of their skin condition which is fed into a machine learning (ML) model to deduce labelling of acne or nothing at all (clear skin WOW)!

💻 How we built it

As first-timers to React Native, we needed to set up both of our laptop workspaces to develop, test, commit & push, merge and pull the many code components of our app. This included downloading a ton of packages and libraries, which was just the start of installing code extensions during our project! In addition to this, we set up our mobile devices with Expo Go to show us the results of our code during runtime. We then made the structured layout of our app which features a load screen and a navigation menu with three pages. Roboflow came in handy to train a model using a dataset of skin condition images. We programmed a camera built into our app featuring Expo Camera and a smart way to save it to your phone to upload to our ML model.

🌵 Challenges we ran into

Saving the photo to Android's Gallery and iOS' Photos on our mobile app using code written on our computers was definitely a challenge! Another dauting problem that we ultimately resolved was successfully feeding the photo taken into our Roboflow model and displaying relevant and defined results on the Results tab.

🏆 Accomplishments that we're proud of

When we applied to this hackathon, we aimed to pour our hearts into creating a project that we could be proud of and that would captivate others as well! Both of us know that we exceeded our expectations on all facets of the project. The React Native layout and UI/UX (as relatively new React Native developers), the functionalities (taking photos using a camera inside of the app, saving & uploading the photo, inputting it through our Roboflow model, and more) using React Native libraries, and displaying helpful and in-depth results for the user to go off of were all huge wina in of themselves!

📚 What we learned

A lesson that we both learned was that making an app is not easy and a streamlined process straight to the finish line. Yet, that is the beauty of making such a product. In the progression of continuosly constructing our app and challenging our limits, we grow and learn as developers and problem solvers alike. We learned that no matter the doubt that one or two people have, the optimism and "dream big" mentality must persevere.

😎✨ What's next for scancare

We eargerly strive to incorporate a feature that allows users to give consent for us as owners of the app to use their photo of their face to expand on our model/dataset to train to be more accurate as time progresses when the app is deployed (another notion that we are working on)!

Built With

• expo-go
• javascript
• kotlin
• react-native
• roboflow
• typescript