ICIP 2017 paper written by Philipp Werner, Frerk Saxen, and Ayoub Al-Hamadi

Abstract

Head pose estimation can help in understanding human behavior or to improve head pose invariance in various face analysis applications. Ready-to-use pose estimators are available with several facial landmark trackers, but their accuracy is commonly unknown. Following the goal to find the best landmark based pose estimator, we introduce a new database (called SyLaHP), propose a new benchmark protocol, and describe and implement a method to learn a pose estimator on top of any landmark detector (called HPFL). The experiments (including cross database) reveal that OpenFace comes with the best pose estimator. Further, HPFL models trained on top of landmark trackers outperform the respective built-in pose estimators. The SyLaHP database, source code, and trained models are publicly available for research.

The paper PDF is available here. A poster is also available here.

Downloads

The SyLaHP dataset (797 MB) and the Matlab source code including trained models (171 MB) are available for non-commercial research purposes. If you use it for any publication, please cite the following paper:

Philipp Werner, Frerk Saxen, and Ayoub Al-Hamadi,
"Landmark based Head Pose Estimation Benchmark and Method",
International Conference on Image Processing (ICIP), 2017.