Abstract-Analysis of a driver's head behavior is an integral part of a driver monitoring system. In particular, the head pose and dynamics are strong indicators of a driver's focus of attention. Many existing state-of-the-art head dynamic analyzers are, however, limited to single-camera perspectives, which are susceptible to occlusion of facial features from spatially large head movements away from the frontal pose. Nonfrontal glances away from the road ahead, however, are of special interest since interesting events, which are critical to driver safety, occur during those times. In this paper, we present a distributed camera framework for head movement analysis, with emphasis on the ability to robustly and continuously operate even during large head movements. The proposed system tracks facial features and analyzes their geometric configuration to estimate the head pose using a 3-D model. We present two such solutions that additionally exploit the constraints that are present in a driving context and video data to improve tracking accuracy and computation time. Furthermore, we conduct a thorough comparative study with different camera configurations. For experimental evaluations, we collected a novel head pose data set from naturalistic on-road driving in urban streets and freeways, with particular emphasis on events inducing spatially large head movements (e.g., merge and lane change). Our analyses show promising results.

Abstract-Analysis of driver's head behavior is an integral part of driver monitoring system. Driver's coarse gaze direction or gaze zone is a very important cue in understanding driverstate. Many existing gaze zone estimators are, however, limited to single camera perspectives, which are vulnerable to occlusions of facial features from spatially large head movements away from the frontal pose. Non-frontal glances away from the driving direction, though, are of special interest as interesting events, critical to driver safety, occur during those times. In this paper, we present a distributed camera framework for gaze zone estimation using head pose dynamics to operate robustly and continuously even during large head movements. For experimental evaluations, we collected a dataset from naturalistic on-road driving in urban streets and freeways. A human expert provided the gaze zone ground truth using all vision information including eyes and surround context. Our emphasis is to understand the efficacy of the head pose dynamic information in predicting eye-gaze-based zone ground truth. We conducted several experiments in designing the dynamic features and compared the performance against static head pose based approach. Analyses show that dynamic information significantly improves the results.

Abstract-Driver's gaze direction is a critical information in understanding driver state. In this paper, we present a distributed camera framework to estimate driver's coarse gaze direction using both head and eye cues. Coarse gaze direction is often sufficient in a number of applications, however, the challenge is to estimate gaze direction robustly in naturalistic realworld driving. Towards this end, we propose gaze-surrogate features estimated from eye region via eyelid and iris analysis. We present a novel iris detection computational framework. We are able to extract proposed features robustly and determine driver's gaze zone effectively. We evaluated the proposed system on a dataset, collected from naturalistic on-road driving in urban streets and freeways. A human expert annotated driver's gaze zone ground truth using information from the driver's eyes and the surrounding context. We conducted two experiments to compare the performance of the gaze zone estimation with and without eye cues. The head-alone experiment has a reasonably good result for most of the gaze zones with an overall 79.8% of weighted accuracy. By adding eye cues, the experimental result shows that the overall weighted accuracy is boosted to 94.9%, and all the individual gaze zones have a better true detection rate especially between the adjacent zones. Therefore, our experimental evaluations show efficacy of the proposed features and very promising results for robust gaze zone estimation.

No abstract