ACM SIGBED Review

Eckert, Marie; Robotham, Thomas; Habets, Emanuël A. P.; Rummukainen, Olli S.

doi: 10.1145/3530798pmid: N/A

Virtual reality (VR) headsets with an integrated eye tracker enable the measurement of pupil size fluctuations correlated with cognition during a VR experience. We present a method to correct for the light-induced pupil size changes, otherwise masking the more subtle cognitively-driven effects, such as cognitive load and emotional state. We explore multiple calibration sequences to find individual mapping functions relating the luminance to pupil dilation that can be employed in real-time during a VR experience. The resulting mapping functions are evaluated in a VR-based n-back task and in free exploration of a six-degrees-of-freedom VR scene. Our results show estimating luminance from a weighted average of the fixation area and the background yields the best performance. Calibration sequence composed of either solid gray or realistic scene brightness levels shown for 6 s in a pseudo-random order proved most robust.

Stoeve, Maike; Wirth, Markus; Farlock, Rosanna; Antunovic, André; Müller, Victoria; Eskofier, Bjoern M.

doi: 10.1145/3530796pmid: N/A

Monitoring stress is relevant in many areas, including sports science. In that scope, various studies showed the feasibility of stress classification using eye tracking data. In most cases, the screen-based experimental design restricted the motion of participants. Consequently, the transferability of results to dynamic sports applications remains unclear. To address this research gap, we conducted a virtual reality-based stress test consisting of a football goalkeeping scenario. We contribute by proposing a stress classification pipeline solely relying on gaze behaviour and pupil diameter metrics extracted from the recorded data. To optimize the analysis pipeline, we applied feature selection and compared the performance of different classification methods. Results show that the Random Forest classifier achieves the best performance with 87.3% accuracy, comparable to state-of-the-art approaches fusing eye tracking data and additional biosignals. Moreover, our approach outperforms existing methods exclusively relying on eye measures.

Koch, Maurice; Weiskopf, Daniel; Kurzhals, Kuno

doi: 10.1145/3530795pmid: N/A

Comparing mobile eye tracking data from multiple participants without information about areas of interest (AOIs) is challenging because of individual timing and coordinate systems. We present a technique, the gaze spiral, that visualizes individual recordings based on image content of the stimulus. The spiral layout of the slitscan visualization is used to create a compact representation of scanpaths. The visualization provides an overview of multiple recordings even for long time spans and helps identify and annotate recurring patterns within recordings. The gaze spirals can also serve as glyphs that can be projected to 2D space based on established scanpath metrics in order to interpret the metrics and identify groups of similar viewing behavior. We present examples based on two egocentric datasets to demonstrate the effectiveness of our approach for annotation and comparison tasks. Our examples show that the technique has the potential to let users compare even long-term recordings of pervasive scenarios without manual annotation.

Kaur, Harsimran; Jindal, Swati; Manduchi, Roberto

doi: 10.1145/3530797pmid: 35754936

Over the past several years, a number of data-driven gaze tracking algorithms have been proposed, which have been shown to outperform classic model-based methods in terms of gaze direction accuracy. These algorithms leverage the recent development of sophisticated CNN architectures, as well as the availability of large gaze datasets captured under various conditions. One shortcoming of black-box, end-to-end methods, though, is that any unexpected behaviors are difficult to explain. In addition, there is always the risk that a system trained with a certain dataset may not perform well when tested on data from a different source (the "domain gap" problem.) In this work, we propose a novel method to embed eye geometry information in an end-to-end gaze estimation network by means of a "geometric layer". Our experimental results show that our system outperforms other state-of-the-art methods in cross-dataset evaluation, while producing competitive performance over within dataset tests. In addition, the proposed system is able to extrapolate gaze angles outside the range of those considered in the training data.