I am looking for an efficient head pose estimation algorithm from facial features.

What I already have: I already have implemented the code for stable facial landmarks(68 landmarks on the face). I also have the intrinsic camera parameters.

The task is to efficiently determine the head pose from webcam video i.e, (roll, yaw and pitch).

Which are the viable algorithms for the purpose?

I am working on python by the way, any working code for the same would be welcome.

  • $\begingroup$ unless you absolutely have to use your implementation, there is already a very good implementation to find head position for Kinect for Windows in C# that i used in my thesis (:. without a kinect you may be able to dig through the C# SDK to find how they do it, although it will be heavily abstracted $\endgroup$ – panthyon Oct 13 '15 at 13:43
  • $\begingroup$ @panthyon I cannot use kinect, have to implement it using webcam, and since I already have facial landmarks I am looking for a geometric method that can do the task, preferably in python $\endgroup$ – Amal Vincent Oct 14 '15 at 4:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.