VIVA Face-off Challenge: Dataset Creation and Balancing Privacy

Vision for intelligent vehicles is a growing area of research [5] for many practical reasons including the relatively inexpensive nature of camera sensing units and even more the non-contact and non-intrusive manner of observation. The latter is of critical importance when observing the driver inside the vehicle cockpit because no sensing unit should impede the driver’s primary task of driving. One of the key tasks in observing the driver is to estimate the driver’s gaze direction. From a vision sensing perspective, for driver gaze estimation, two of the fundamental building blocks are face detection and head pose estimation.