Evaluating the Tobii Pro Glasses 2 and 3 in static and dynamic conditions

Abstract

Over the past few decades, there have been significant developments in eye-tracking technology, particularly in the domain of mobile, head-mounted devices. Nevertheless, questions remain regarding the accuracy of these eye-trackers during static and dynamic tasks. In light of this, we evaluated the performance of two widely used devices: Tobii Pro Glasses 2 and Tobii Pro Glasses 3. A total of 36 participants engaged in tasks under three dynamicity conditions. In the “seated with a chinrest” trial, only the eyes could be moved; in the “seated without a chinrest” trial, both the head and the eyes were free to move; and during the walking trial, participants walked along a straight path. During the seated trials, participants’ gaze was directed towards dots on a wall by means of audio instructions, whereas in the walking trial, participants maintained their gaze on a bullseye while walking towards it. Eye-tracker accuracy was determined using computer vision techniques to identify the target within the scene camera image. The findings showed that Tobii 3 outperformed Tobii 2 in terms of accuracy during the walking trials. Moreover, the results suggest that employing a chinrest in the case of head-mounted eye-trackers is counterproductive, as it necessitates larger eye eccentricities for target fixation, thereby compromising accuracy compared to not using a chinrest, which allows for head movement. Lastly, it was found that participants who reported higher workload demonstrated poorer eye-tracking accuracy. The current findings may be useful in the design of experiments that involve head-mounted eye-trackers.

Video-Based Gaze Detection for Oculomotor Abnormality Measurements

“…Compared to the use of a specialized IR eye tracker as a reference ground truth, we found that the NLGEM alone achieved an absolute visual angle accuracy (MAE) of 4.80 degrees. While such an accuracy level is comparable to that of some of the leading IR eye trackers available at present [23,24] and may be sufficient for applications that require general visual attention direction or saliency distribution on display, most eye tracking applications require higher accuracy, especially in medical applications (e.g., for the measurement of eye movements in patients with various diseases and oculomotor symptoms).…”

“…The most commonly used eye trackers at present are Infrared (IR) active sensors, which track the eye's position and orientation to estimate the subject's PoR. These specialized IR eye trackers are currently considered to be the most accurate measurement tool for gaze detection and can be mounted next to a desktop monitor [23] or embedded in wearable head devices (e.g., glasses) [24]. However, even in state-of-the-art IR eye trackers, researchers have observed significant sensitivity to the experimental setup and the diverse characteristics of subjects [25].…”

Less Is More: Higher Skilled Sim Racers Allocate Significantly Less Attention to Track Relative to Display Features Than Lower Skilled Sim Racers

“…They then were given the eye tracking glasses to wear and a one-point calibration was performed. This one-point calibration has been found to be accurate to within 1.60˚ (Onkhar et al, 2023), where accuracy refers to the angular distance from a target.…”

Less Is More: Higher-Skilled Sim Racers Allocate Significantly Less Attention to the Track Relative to the Display Features than Lower-Skilled Sim Racers

“…They were then given the eye tracking glasses to wear and a one-point calibration was performed. This one-point calibration has been found to be accurate to within 1.60° [ 23 ], where accuracy refers to the angular distance from a target.…”