Here is the iBUG head pose estimator. You can use it to estimate head pose in terms of pitch, yaw, and roll from the coordinates of 49 facial landmarks. The algorithm is based on linear regression and rigid 2D alignment, so the code runs quite fasts, which typically takes no more than 1ms to process a frame.
- Install numpy if it's not already there:
pip install numpyorconda install -c anaconda numpy - Copy this package to where Python can see. You don't need to copy test_data though unless you want to run the built-in test.
Please refer to the test function in ibug_head_pose_estimator.py, or just do as follows:
from ibug_head_pose_estimator import HeadPoseEstimator
estimator = HeadPoseEstimator()
pitch, yaw, roll = estimator.estimate_head_pose(landmarks) # landmarks must be a 49x2 matrix