Problem
Research questions and friction points this paper is trying to address.
wearable gaze tracking
driving
gaze accuracy
bird's-eye-view
error quantification
From Laboratory to Road: Evaluating Wearable Gaze Accuracy for Driving
🤖 AI Summary
This study addresses the spatial output deviations of wearable eye trackers in real-world driving caused by head movements and illumination changes, which undermine their reliability as behavioral signals for autonomous driving. To this end, this work proposes the first unified framework to quantify gaze accuracy. By integrating 41 on-road validation scenarios with controlled indoor experiments, it systematically reveals the limitations of offline calibration and demonstrates the necessity of online recalibration and condition-dependent uncertainty estimation. Applying an indoor recording-based offset correction strategy significantly reduces the mean gaze error from 4.58° to 1.10°, comprehensively improving performance across all scenarios. This research fills a critical gap in the reliability assessment of wearable eye tracking for autonomous driving applications.
📝 Abstract
Bird's-eye-view (BEV) representations have become a widely used interface between perception and planning in autonomous driving, but they encode what is in a scene, not what is behaviorally relevant to a human driver. Gaze offers a compelling behavioral signal for this gap, yet wearable eye trackers are routinely deployed as if their spatial output were ground truth, despite known sensitivity to head motion, illumination, and calibration drift. We present, to our knowledge, the first unified framework for quantifying wearable gaze accuracy under real driving conditions. Our on-road study contains 41 validated scenes in which one driver fixated a vehicle's license plate. Gaze error is measured as the angular difference between the plate center and the gaze direction estimated by the glasses. Separate indoor studies with the same driver and device systematically analyze how distance, illumination, head motion, target motion, and gaze eccentricity affect both systematic bias and gaze precision. The mean on-road error was 4.58 degrees. Applying an offset estimated from the indoor recordings reduced it to 1.10 degrees and improved all 41 scenes. Because this offset varied between sessions, reliable BEV supervision may require online recalibration and condition-dependent estimates of gaze uncertainty.
Innovation
Methods, ideas, or system contributions that make the work stand out.
wearable gaze tracking
bird's-eye-view (BEV)
driving evaluation
gaze accuracy
online recalibration
🔎 Similar Papers
No similar papers found.
