Biternion Nets: Continuous Head Pose Regression from Discrete Training Labels

While head pose estimation has been studied for some time, continuous head pose estimation is still an open problem. Most approaches either cannot deal with the periodicity of angular data or require very fine-grained regression labels. We introduce biternion nets, a CNN-based approach that can be trained on very coarse regression labels and still estimate fully continuous $${360}^{\circ }$$ head poses. We show state-of-the-art results on several publicly available datasets. Finally, we demonstrate how easy it is to record and annotate a new dataset with coarse orientation labels in order to obtain continuous head pose estimates using our biternion nets.