/FaceMe

Assisting individuals with Prosopagnosia using computer vision and label detection. Part of AstraZeneca's Neurodiversity Hackathon.

Primary LanguageTypeScript

Faceme - AstraZeneca Neurodiversity Hack

Assisting individuals with Prosopagnosia using label detection and facial recognition

Demo Video

Individuals with Prosopagnosia often experience that "it's on the tip of my tongue!" feeling when they can't connect a name to a face and have to focus on anothers facial feautures to remember them. Faceme's goal is remedy this issue. With the power of AI, when a user uploads an image of a person, the app provides a set of facial features for the image and then prompts the user to give a name. Then users can filter saved images by searching by features and can view the name of the person.

Website Architecture

Backend

  • Python & Flask
    • Authentication with JSON Web Tokens
  • MongoDB
  • Firebase
    • Image storage
  • Google Cloud Vision API

Frontend

  • TypeScript + React
  • Redux State Management
  • Material UI
  • Yarn

Setup

Backend

  • Create a virtual environment and download the necessary dependencies by running: pip install -r requirements.txt

  • Setup a mongodb account and create a .env file to store the uri in.

  • Then start the server with:python app.py

Frontend

  • If you don't have yarn installed run: npm install --global yarn

  • Run yarn install to download the dependencies

  • Run yarn start to start the frontend web server