Are You Braden?

My friend Braden, who I became good friends with at Cal Poly, has always been fascinated with computer science, even though his major, economics, had little to do with it. One particular thing Braden always was amazed by was this app on his phone that would scan a picture and tell you whether or not a hot dog was in it. This silly app and Braden's fixation on it inspired me to spend a couple days writing something similar in hopes of blowing his mind!

I got to work, researching different facial recognitions method. Due to the fact I wanted this to be simple, lightweight, and fast, I decided I did not want to go down the machine learning route. Instead, I discovered several packages on python, one called "facial-recognition" that used the HOG method to perform facial recognition. After learning more about the HOG method, I gathered several pictures of Braden, detected his face, and saved it as an encoding. I then created a simple flask backend that would receive images from requests and use this encodings file to determine whether or not the image that it received included Braden. The Flask backend would send back a new image with the faces boxed (if there was any) and would let the sender know whether or not it was Braden.

The last thing to do was create an interface where users could actually take photos of themselves. I did this by creating a simple React frontend that allows users to take a photo of themselves and send it to the backend. The frontend will then render a positive or negative message based on the data received from the backend.

This project is imperfect, and definitely will mess up on the occasion. However, I found this to be a great learning experience, as I learned about a new facial recognition (HOG) and Braden ended up loving the application. So in my opinion, it was a great success!