This repo shows how to estimate human head pose from videos using TensorFlow and OpenCV.
The following packages are required:
- TensorFlow 1.14.
- OpenCV 3.3 or higher.
- Python 3.5
The code is tested on Ubuntu 16.04.
This repository comes with a pre-trained model for facial landmark detection. Just git clone then you are good to go.
# From your favorite development directory:
git clone https://github.com/yinguobing/head-pose-estimation.git
A video file or a webcam index should be assigned through arguments. If no source provided, the default webcam will be used.
For any video format that OpenCV supported (mp4
, avi
etc.):
python3 estimate_head_pose.py --video /path/to/video.mp4
The webcam index should be assigned:
python3 estimate_head_pose.py --cam 0
There are three major steps:
-
Face detection. A face detector is adopted to provide a face box containing a human face. Then the face box is expanded and transformed to a square to suit the needs of later steps.
-
Facial landmark detection. A custom trained facial landmark detector based on TensorFlow is responsible for output 68 facial landmarks.
-
Pose estimation. Once we got the 68 facial landmarks, a mutual PnP algorithms is adopted to calculate the pose.
The marks is detected frame by frame, which result in small variance between adjacent frames. This makes the pose unstable. A Kalman filter is used to solve this problem, you can draw the original pose to observe the difference.
To reproduce the facial landmark detection model, you can refer to this series of posts(in Chinese only). And the training code is also open sourced: https://github.com/yinguobing/cnn-facial-landmark
This project is licensed under the MIT License - see the LICENSE.md file for details
Yin Guobing (尹国冰) - yinguobing
The pre-trained TensorFlow model file is trained with various public data sets which have their own licenses. Please refer to them before using this code.
- 300-W: https://ibug.doc.ic.ac.uk/resources/300-W/
- 300-VW: https://ibug.doc.ic.ac.uk/resources/300-VW/
- LFPW: https://neerajkumar.org/databases/lfpw/
- HELEN: http://www.ifp.illinois.edu/~vuongle2/helen/
- AFW: https://www.ics.uci.edu/~xzhu/face/
- IBUG: https://ibug.doc.ic.ac.uk/resources/facial-point-annotations/
The 3D model of face comes from OpenFace, you can find the original file here.
The build in face detector comes from OpenCV. https://github.com/opencv/opencv/tree/master/samples/dnn/face_detector