Analyzing dynamic postures of vehicle occupants in various situations is valuable for improving occupant accommodation and safety. Accurate tracking of an occupant’s head is one of the key factors for estimating occupant dynamic postures in vehicle since the head potentially has a large range of motion, controls gaze, and may require special protection in dynamic events including crashes. Previous vehicle occupant posture studies have primarily used marker-based optical motion capture systems or multiple video cameras for tracking facial features or markers on the head. However, the former approach has a limitation in collecting on-road data, and the latter one is limited by the lack of accuracy without intensive manual postprocessing. This study developed an automated on-road head tracking method using a Microsoft Kinect V2 sensor based on time-of-flight measurement technology. The Kinect depth data of vehicle passengers were recorded during hard braking and rapid lane changes. The dynamic head orientation and location data were obtained by fitting a subject-specific 3D head model to the depth data from each frame. Results were validated using a tracking system based on multiple video images. The results showed that the proposed method and system provides efficient and accurate in-vehicle head tracking using a single low-cost depth camera. Extensions of this method have broad applications for study of vehicle occupant dynamics, and with advances in technology may be applicable to routine use in production vehicles.