FacialRecognitionAdvertising

Displays billboard advertisements to users based off attributes detected via facial recognition software.

Built at RowdyHacks2019 with the help of

@defaultidname: Franklyn Ramirez

@MilesMartinez

@BenGonzalez97 I also worked on the OpenCV library to implement attributes from the webcam feed. We did a bunch of web app development, which I was fairly new to and never dived into before. It was a great learning experience and happy with the end result.

Inspiration

Digital marketing
LED billboards
Facebook Ads Platform

What it does

Obtains user attributes via web cam and tailors ads to identified demographic

How we built it

The OpenCV python library was used for facial detection
Neural net models detect user attributes set by developer
Advertisements were then shown based off of the user attributes detected

Challenges we ran into

Integrating the facebook graph API
Interfacing python with Node.js
Parsing JSON data from python requests module

Accomplishments that we're proud of

Displaying ads based on user profile/attributes

What we learned

HTTP requests
Complex API integrations
Deploying a neural network

What's next for FacialRecognitionAdvertising

Detecting more user attributes
Increasing accuracy of attribute classification
Integrating data with advertising API

ungtb10d/FacialRecognitionAdvertising