Catadioptric vision system for an optimal observation of the driver face and the road scene

In this paper, we propose to design a new compact sensor for the simultaneous monitoring of the driver activity and the road scene. This sensor will be integrated in a driver assistant system to study the correlation between the driver’s gaze and the road scene. The device is based on a catadioptric configuration combining two different reflective surfaces. One enables the capture of a panoramic view of the environment in and out of the vehicle. Another has been designed to obtain a sufficient resolution for the eye tracking of the driver. With this new sensor, a 3D reconstruction by stereovision can be computed as two different projections of the driver’s face can be contemplated on the same image. The complete design is validated with the study of the accuracy of stereovision reconstruction.