Home > Research > Publications & Outputs > Gaussian processes as an alternative to polynom...

Links

Text available via DOI:

View graph of relations

Gaussian processes as an alternative to polynomial gaze estimation functions

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Gaussian processes as an alternative to polynomial gaze estimation functions. / Sesma-Sanchez, Laura; Zhang, Yanxia; Bulling, Andreas; Gellersen, Hans-Werner Georg.

ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications. New York : ACM, 2016. p. 229-232.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Sesma-Sanchez, L, Zhang, Y, Bulling, A & Gellersen, H-WG 2016, Gaussian processes as an alternative to polynomial gaze estimation functions. in ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications. ACM, New York, pp. 229-232. https://doi.org/10.1145/2857491.2857509

APA

Sesma-Sanchez, L., Zhang, Y., Bulling, A., & Gellersen, H-W. G. (2016). Gaussian processes as an alternative to polynomial gaze estimation functions. In ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications (pp. 229-232). ACM. https://doi.org/10.1145/2857491.2857509

Vancouver

Sesma-Sanchez L, Zhang Y, Bulling A, Gellersen H-WG. Gaussian processes as an alternative to polynomial gaze estimation functions. In ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications. New York: ACM. 2016. p. 229-232 https://doi.org/10.1145/2857491.2857509

Author

Sesma-Sanchez, Laura ; Zhang, Yanxia ; Bulling, Andreas ; Gellersen, Hans-Werner Georg. / Gaussian processes as an alternative to polynomial gaze estimation functions. ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications. New York : ACM, 2016. pp. 229-232

Bibtex

@inproceedings{7adbe0d1ce304905a6ddc1f6e4a00206,
title = "Gaussian processes as an alternative to polynomial gaze estimation functions",
abstract = "Interpolation-based methods are widely used for gaze estimation due to their simplicity. In particular, feature-based methods that map the image eye features to gaze, are very popular. The most spread regression function used in this kind of method is the polynomial regression. In this paper, we present an alternative regression function to estimate gaze: the Gaussian regression. We show how the Gaussian processes can better adapt to the non-linear behavior of the eye movement, providing higher gaze estimation accuracies. The Gaussian regression is compared, in a simulated environment, to the polynomial regression, when using the same mapping features, the normalized pupil center-corneal reflection and pupil center-eye corners vectors. This comparison is done for three different screen sizes. The results show that for larger screens, where wider gaze angles are required, i.e., the non-linear behavior of the eye is more present, the outperformance of the Gaussian regression is more evident. Furthermore, we can conclude that, for both types of regressions, the gaze estimation accuracy increases for smaller screens, where the eye movements are more linear.",
keywords = "gaze estimation, Gaussian process, polynomials",
author = "Laura Sesma-Sanchez and Yanxia Zhang and Andreas Bulling and Gellersen, {Hans-Werner Georg}",
year = "2016",
month = mar,
day = "14",
doi = "10.1145/2857491.2857509",
language = "English",
isbn = "9781450341257",
pages = "229--232",
booktitle = "ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications",
publisher = "ACM",

}

RIS

TY - GEN

T1 - Gaussian processes as an alternative to polynomial gaze estimation functions

AU - Sesma-Sanchez, Laura

AU - Zhang, Yanxia

AU - Bulling, Andreas

AU - Gellersen, Hans-Werner Georg

PY - 2016/3/14

Y1 - 2016/3/14

N2 - Interpolation-based methods are widely used for gaze estimation due to their simplicity. In particular, feature-based methods that map the image eye features to gaze, are very popular. The most spread regression function used in this kind of method is the polynomial regression. In this paper, we present an alternative regression function to estimate gaze: the Gaussian regression. We show how the Gaussian processes can better adapt to the non-linear behavior of the eye movement, providing higher gaze estimation accuracies. The Gaussian regression is compared, in a simulated environment, to the polynomial regression, when using the same mapping features, the normalized pupil center-corneal reflection and pupil center-eye corners vectors. This comparison is done for three different screen sizes. The results show that for larger screens, where wider gaze angles are required, i.e., the non-linear behavior of the eye is more present, the outperformance of the Gaussian regression is more evident. Furthermore, we can conclude that, for both types of regressions, the gaze estimation accuracy increases for smaller screens, where the eye movements are more linear.

AB - Interpolation-based methods are widely used for gaze estimation due to their simplicity. In particular, feature-based methods that map the image eye features to gaze, are very popular. The most spread regression function used in this kind of method is the polynomial regression. In this paper, we present an alternative regression function to estimate gaze: the Gaussian regression. We show how the Gaussian processes can better adapt to the non-linear behavior of the eye movement, providing higher gaze estimation accuracies. The Gaussian regression is compared, in a simulated environment, to the polynomial regression, when using the same mapping features, the normalized pupil center-corneal reflection and pupil center-eye corners vectors. This comparison is done for three different screen sizes. The results show that for larger screens, where wider gaze angles are required, i.e., the non-linear behavior of the eye is more present, the outperformance of the Gaussian regression is more evident. Furthermore, we can conclude that, for both types of regressions, the gaze estimation accuracy increases for smaller screens, where the eye movements are more linear.

KW - gaze estimation

KW - Gaussian process

KW - polynomials

U2 - 10.1145/2857491.2857509

DO - 10.1145/2857491.2857509

M3 - Conference contribution/Paper

SN - 9781450341257

SP - 229

EP - 232

BT - ETRA '16 Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications

PB - ACM

CY - New York

ER -