A Differential 2D Gaussian Ellipse-Based Eye Movement Analysis

Abstract

As data-driven analysis methods powered by artificial intelligence have matured, research on visual attention prediction has advanced markedly. However, gaze-point data from eye-tracking devices are often characterized by high noise levels, limiting accuracy in representing real driver behavior. Thus, we proposed the Differential 2D Gaussian Ellipse (D2DGE) representation, which captures gaze distribution within a time window to reduce noise from devices or unconscious glances. To validate D2DGE, a Generative Adversarial Imitation Learning model was trained on both raw gaze data and the D2DGE data, and we assessed the similarity between the raw and generated data using the Kullback–Leibler divergence. Then, we compared the five parameters of D2DGE representation between novice and experienced drivers. Results show that the D2DGE data can better approximate the raw data and contains richer information than raw gaze data. The findings indicate that the D2DGE can be a promising alternative to describe gaze distribution during driving.

Keywords

attention human factors and behavior modeling and analytics eye-tracking

Get full access to this article

View all access options for this article.

References

Ahlström

Kircher

Nyström

Wolfe

(2021). Eye tracking in driver attention research - How gaze data interpretations influence what we learn. Frontiers in Neuroergonomics, 2, Article 778043.

Baee

Pakdamanian

Kim

Feng

Ordonez

Barnes

(2021). Medirl: Predicting the visual attention of drivers via maximum entropy deep inverse reinforcement learning [Conference session]. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Korea, pp. 13178–13188.

Corbeil

R. R.

Searle

S. R.

(1976). Restricted maximum likelihood (REML) estimation of variance components in the mixed model. Technometrics, 18(1), 31–38.

Engineering Systems Technologies (EST) GmbH & Co. KG. (2025). DIKABLIS GLASSES 3: High precision eye tracking technology [Product webpage]. Retrieved May 3, 2025, from https://www.est-kl.com/zh-hans/dikablis-glasses-3/

Huang

Sun

Chen

(2023). A multimodal deep neural network for prediction of the driver’s focus of attention based on anthropomorphic attention mechanism and prior knowledge. Expert Systems with Applications, 214, Article 119157.

Gómez-Poveda

Gaudioso

(2016). Evaluation of temporal stability of eye tracking algorithms using webcams. Expert Systems With Applications, 64, 69–83.

(2020). Understanding and supporting anticipatory driving in automated vehicles. University of Toronto (Canada).

Hershey

J. R.

Olsen

P. A.

(2007, April). Approximating the Kullback-Leibler divergence between Gaussian mixture models. In 2007 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (Vol. 4, pp. IV–317). IEEE.

Ermon

(2016). Generative adversarial imitation learning [Conference session]. Proceedings of the 30th International Conference on Neural Information Processing Systems, Changsha, China, pp. 4572–4580.

10.

Larson

A. M.

Loschky

L. C.

(2009). The contributions of central versus peripheral vision to scene gist recognition. Journal of Vision, 9(10), 6.

11.

Schulman

Wolski

Dhariwal

Radford

Klimov

(2017). Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347.

12.

Sivak

(1996). The information that drivers use: Is it indeed 90% visual? Perception, 25(9), 1081–1089.

13.

Smart Eye. (2025). Smart Eye Pro: Remote eye tracking for human behavior research [Product webpage]. Smart Eye AB. Retrieved May 2, 2025, from https://www.smarteye.se/smart-eye-pro/

14.

Tobii

. (2025). Tobii Pro Glasses 3: Specifications [Product webpage]. Retrieved May 3, 2025, from https://www.tobii.com/products/eye-trackers/wearables/tobii-pro-glasses-3#specifications

15.

Underwood

(2007). Visual attention and the transition from novice to advanced driver. Ergonomics, 50(8), 1235–1249.

16.

Vater

Wolfe

Rosenholtz

(2022). Peripheral vision in real-world tasks: A systematic review. Psychonomic Bulletin & Review, 29(5), 1531–1557.

17.

Wang

Kim

Wenxuan

Xie

Chen

Luo

(2024). DeepAccident: A motion and accident prediction benchmark for V2X autonomous driving. Proceedings of the AAAI Conference on Artificial Intelligence, 38(6), 5599–5606.

18.

Wang

Xie

Dong

Shan

(2021). Real-ESRGAN: Training real-world blind super-resolution with pure synthetic data [Conference session]. Proceedings of the IEEE/CVF International Conference on Computer Vision, Montreal, QC, Canada, pp. 1905–1914.

19.

Werneke

Vollrath

(2012). What does the driver look at? The influence of intersection characteristics on attention allocation and driving behavior. Accident Analysis & Prevention, 45, 610–619.

20.

Wolfe

Dobres

Rosenholtz

Reimer

(2017). More than the useful field: Considering peripheral vision in driving. Applied Ergonomics, 65, 316–325.

21.

Tao

Wang

Liu

Zheng

(2024). Perception-oriented video frame interpolation via asymmetric blending [Conference session]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville TN, USA, pp. 2753–2762.