Driver Gaze Region Estimation Without Using Eye Movement
This work addresses a critical need for robust driver attention monitoring in Advanced Driver Assistance Systems, though it is incremental by focusing on region-based estimation instead of fine-grained gaze tracking.
The paper tackled the problem of estimating a driver's gaze region without eye movement by using facial features and head pose, achieving an average accuracy of 91.4% at 11 Hz on a dataset of 50 drivers.
Automated estimation of the allocation of a driver's visual attention may be a critical component of future Advanced Driver Assistance Systems. In theory, vision-based tracking of the eye can provide a good estimate of gaze location. In practice, eye tracking from video is challenging because of sunglasses, eyeglass reflections, lighting conditions, occlusions, motion blur, and other factors. Estimation of head pose, on the other hand, is robust to many of these effects, but cannot provide as fine-grained of a resolution in localizing the gaze. However, for the purpose of keeping the driver safe, it is sufficient to partition gaze into regions. In this effort, we propose a system that extracts facial features and classifies their spatial configuration into six regions in real-time. Our proposed method achieves an average accuracy of 91.4% at an average decision rate of 11 Hz on a dataset of 50 drivers from an on-road study.