Inspiration
I was tired of having to choose between phone numbers, emails, and social media accounts to give to my associates, so I decided access for each person to your contact info should be easy.
What it does
When you meet someone else in public with the app and want to share contact info, all you have to do is snap a pic of them. It'll send them a notification, and then each of you can share what specific phone numbers, emails, social media and other info that you want the other to see. It also keeps a log of when and where you met them, and keeps the pic you two snapped together.
How I built it
I used OpenCV, Python, and Postgresql to create a backend that could identify faces and store images, and Flutter to create an iOS and Android mobile app.
Challenges I ran into
Time, like for most people, was my biggest issue. Unfortunately, the small things can hurt the most. While it only took me about twenty minutes to get the server to recognize known faces, it took me the first three hours of the hackathon to accept the Android Studio licenses in order to develop the app.
(Here's the link to the backend - https://github.com/dannyhyatt/hackathon-nmeet-backend)
Accomplishments that I'm proud of
I'm proud that I was able to turn my first facial recognition project into a functioning app that can recognize faces. While I didn't complete the app entirely, I finished all of the facial recognition components and I know completing the rest of the app is trivial.
What I learned
I learned how accurate facial recognition has recently become - and how it still has significant errors. I actually implemented a part of my program that uses location to provide an accurate facial match with many people.
What's next for nMeet
I plan on finishing it and polishing it up, and maybe publishing it as a concept app. I don't plan to make this functional, but I think it would be a cool part of Snapchat or Instagram and may contact them with a feature suggestion and my app as demo.
Log in or sign up for Devpost to join the conversation.