Android applications for a facial recognition systems using the MOD LIVE HUD.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.

#facerecHUD ###WINLAB Summer Internship Project

There are three pieces to this project: the Passthrough and Output apps in the repository, and the server.

The idea of this project is to create a Head-Up Display system that will recognize faces and display relevant information.

Software Design


This is the Android app that runs on the HUD device, in this case a Recon MOD LIVE HUD, displays text on the HUD. It's essentially the SDK demo on Bluetooth Chat, with an interface modified for the small screen.


This is the Android app that runs on a smartphone. This is necessary because the MOD LIVE doesn't have its own camera or networking, so switching to something like Google Glass would eliminate this need.


This is the Flask server that runs the Fischerfaces algorithm for facial detection. This is offloaded to server because of potentially expensive computations, but could be incorporated into the main Android app.

Software Flow

  • HUD detects a keypress on Bluetooth remote, forwards it to the phone.
  • Phone takes a picture, and if a face is detected, crops to it and sends it to the server.
  • Server runs recognition algorithm, sends results to phone.
  • Phone looks up result in information database, sends information to HUD.
  • HUD displays information. There is much rejoicing.