Facial recognition and tracking system with the Raspberry Pi and RPiSoC

A Face tracking system developed on the Raspberry pi, using the RPi camera module, and realized using OpenCV. The identified faces are followed using a pan/tilt mechanism controlled by two servomotors. The servomotor control is offloaded to an RPiSoC using SPI. Check it out on youtube!

face tracking

You can easily extend this project to suit all your computer vision needs, so stay tuned because we intend to soon post all source code as well as ideas for extending the project. We’ll even give a little introductory insight into the mathematics of facial recognition. So keep an eye on this blog!

Comments (7)

  1. Ram kumar (reply)

    April 10, 2015 at 11:02 pm

    Hi Team,
    I want to purchase your project Facial recognition and tracking system with the Raspberry Pi and RPiSoC for my under graduate project.could you send details and cost of that project.it help me to learn more to do project.I am from India.reply as soon as possible.

    Thanks in advance,

    Ramkumar v

    1. admin (reply)

      April 28, 2015 at 7:36 pm

      Hi Ram,

      We are actually planning to launch a kickstarter in the next few weeks and the face tracker will once again be available. However, if you need something much sooner I would suggest looking into using a USB webcam with a Raspberry Pi 2 or Higher spec computer. Our facial recognition software isn’t the unique part in the project and similar examples can be found online. We found a way to make the Raspberry Pi camera work with OpenCV, allowing for a much higher performance computer vision system on the pi than previously thought possible.


  2. ahmed (reply)

    June 9, 2016 at 9:10 am

    the code plz

    1. Robert Barron (reply)

      June 16, 2016 at 10:04 pm

      Hi Ahmed,

      The code has now been posted to our github here: https://github.com/EmbeditElectronics/Face-Tracker

      We are also working on a full writeup that will explain the entire process of getting your face tracker up and running.


  3. Raj Thakkar (reply)

    October 24, 2016 at 5:24 am

    Nice Blog! Thanks for this helpful information I agree with all points you have given to us. I will follow all of them.-
    Raspberry PI Solutions

  4. David Cooper (reply)

    September 24, 2017 at 7:14 pm

    First, Thank you for this project.
    I took your C code and revamped the tracker portion to work with my i2c pan/tilt servo controller. That seems to work but I am having trouble with the face recognition section of the program. Facial recognition seems to be VERY slow. I do occasionally see a rectangular box around my face but it is very sporadic.
    Can you think of something that I can try to ‘Fix’ my problem.

    Other facial recognition programs seem to identify my face much better and faster.

    My system is a Pi 3 with Strech, a standard Pi camera version 2, OpenCV 3.3. I have not modified any of the other 2 threads. CPU utilization seems to be between 50 and 60%.

Leave a Reply

Your email address will not be published.

Published on: 13 June 2014
Posted by: admin
Discussion: 7 Comments