#OPEN CV EYE TRACKING CODE#
If you want to use the code in this article, please make sure that you have these (or newer) versions. Using the OpenCV library, you can make use of the HAAR cascade filters to do this efficiently.ĭuring the implementation, we made use of Anaconda with Python 3.5, OpenCV 3.1.0, and dlib 19.1.0. Detecting a faceĪfter we decided to make use of Python, the first feature we would need for performing face recognition is to detect where in the current field of vision a face is present. In this blog post, I want to focus on showing how we made use of Python and OpenCV to detect a face and then use the dlib library to efficiently keep tracking the face. In order do this, the first thing we will have to do is to detect faces and keep tracking them. The two components of the robot you cannot see are a laptop (the actual brain of our robot) and a router (connecting the EV3 unit and the iPhone with the laptop).Īs mentioned, one of the features of our robot is that it will do face recognition. Fortunately, we were able to finish everything before the 18th:Īlthough the robot is visible in the picture above, being held by Naser Bakhshi, you cannot see it that well, so below a close-up of the robot we demonstrated at the TEDx:Īs you can see in the picture of the robot, it consists of a Lego Mindstorms EV3 unit, some additional Lego Technic, and an iPhone. The moment I heard about this project, I knew I wanted to be involved :)ĭuring the course of the project, we got asked whether we would be able to have the robot ready before the 18th of November: Deloitte was sponsoring the TEDx in Amsterdam and it would be great to show the robot at the Deloitte stand. At work, I was asked whether I wanted to help out on a project dealing with a robot that could do autonomous navigation and combine this with both speech recognition and most importantly: face recognition.
0 kommentar(er)
