BayesGaze: A bayesian approach to eye-gaze based target selection

Zhi Li; Maozheng Zhao; Yifan Wang; Sina Rashidian; Furqan Baig; Rui Liu; Wanyu Liu; Michel Beaudouin-Lafon; Brooke Ellison; Fusheng Wang; I. V. Ramakrishnan; Xiaojun Bi

doi:10.20380/GI2021.35

BayesGaze: A bayesian approach to eye-gaze based target selection

Zhi Li
, Maozheng Zhao
, Yifan Wang
, Sina Rashidian
, Furqan Baig
, Rui Liu
, Wanyu Liu
, Michel Beaudouin-Lafon
, Brooke Ellison
, Fusheng Wang
, I. V. Ramakrishnan
, Xiaojun Bi

Stony Brook University
IRCAM Centre Pompidou
Université Paris-Saclay

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

15 Scopus citations

Abstract

Selecting targets accurately and quickly with eye-gaze input remains an open research question. In this paper, we introduce BayesGaze, a Bayesian approach of determining the selected target given an eye-gaze trajectory. This approach views each sampling point in an eye-gaze trajectory as a signal for selecting a target. It then uses the Bayes' theorem to calculate the posterior probability of selecting a target given a sampling point, and accumulates the posterior probabilities weighted by sampling interval to determine the selected target. The selection results are fed back to update the prior distribution of targets, which is modeled by a categorical distribution. Our investigation shows that BayesGaze improves target selection accuracy and speed over a dwell-based selection method, and the Center of Gravity Mapping (CM) method. Our research shows that both accumulating posterior and incorporating the prior are effective in improving the performance of eye-gaze based target selection.

Original language	English
Title of host publication	Graphics Interface 2021, Proceedings
Editors	Manolis Savva, Ahmed Sabbir Arif
Publisher	Canadian Information Processing Society
ISBN (Electronic)	9780994786869
DOIs	https://doi.org/10.20380/GI2021.35
State	Published - 2021
Event	Graphics Interface 2021 - Virtual, Online Duration: May 27 2021 → May 28 2021

Publication series

Name	Proceedings - Graphics Interface
Volume	2021-May
ISSN (Print)	0713-5424

Conference

Conference	Graphics Interface 2021
City	Virtual, Online
Period	05/27/21 → 05/28/21

Keywords

HCI design and evaluation methods
Human computer interaction (HCI)
Human computer interaction (HCI)
Human computer interaction (HCI); Human-centered computing
Human-centered computing
Interaction techniques; Human-centered computing
User studies

Access to Document

10.20380/GI2021.35

Cite this

Li, Z., Zhao, M., Wang, Y., Rashidian, S., Baig, F., Liu, R., Liu, W., Beaudouin-Lafon, M., Ellison, B., Wang, F., Ramakrishnan, I. V., & Bi, X. (2021). BayesGaze: A bayesian approach to eye-gaze based target selection. In M. Savva, & A. S. Arif (Eds.), Graphics Interface 2021, Proceedings (Proceedings - Graphics Interface; Vol. 2021-May). Canadian Information Processing Society. https://doi.org/10.20380/GI2021.35

@inproceedings{a2f559fe4144463e882f474242f485b4,

title = "BayesGaze: A bayesian approach to eye-gaze based target selection",

abstract = "Selecting targets accurately and quickly with eye-gaze input remains an open research question. In this paper, we introduce BayesGaze, a Bayesian approach of determining the selected target given an eye-gaze trajectory. This approach views each sampling point in an eye-gaze trajectory as a signal for selecting a target. It then uses the Bayes' theorem to calculate the posterior probability of selecting a target given a sampling point, and accumulates the posterior probabilities weighted by sampling interval to determine the selected target. The selection results are fed back to update the prior distribution of targets, which is modeled by a categorical distribution. Our investigation shows that BayesGaze improves target selection accuracy and speed over a dwell-based selection method, and the Center of Gravity Mapping (CM) method. Our research shows that both accumulating posterior and incorporating the prior are effective in improving the performance of eye-gaze based target selection.",

keywords = "HCI design and evaluation methods, Human computer interaction (HCI), Human computer interaction (HCI), Human computer interaction (HCI); Human-centered computing, Human-centered computing, Interaction techniques; Human-centered computing, User studies",

author = "Zhi Li and Maozheng Zhao and Yifan Wang and Sina Rashidian and Furqan Baig and Rui Liu and Wanyu Liu and Michel Beaudouin-Lafon and Brooke Ellison and Fusheng Wang and Ramakrishnan, \{I. V.\} and Xiaojun Bi",

year = "2021",

doi = "10.20380/GI2021.35",

language = "English",

series = "Proceedings - Graphics Interface",

publisher = "Canadian Information Processing Society",

editor = "Manolis Savva and Arif, \{Ahmed Sabbir\}",

booktitle = "Graphics Interface 2021, Proceedings",

}

Li, Z, Zhao, M, Wang, Y, Rashidian, S, Baig, F, Liu, R, Liu, W, Beaudouin-Lafon, M, Ellison, B, Wang, F , Ramakrishnan, IV & Bi, X 2021, BayesGaze: A bayesian approach to eye-gaze based target selection. in M Savva & AS Arif (eds), Graphics Interface 2021, Proceedings. Proceedings - Graphics Interface, vol. 2021-May, Canadian Information Processing Society, Graphics Interface 2021, Virtual, Online, 05/27/21. https://doi.org/10.20380/GI2021.35

BayesGaze: A bayesian approach to eye-gaze based target selection. / Li, Zhi; Zhao, Maozheng; Wang, Yifan et al.
Graphics Interface 2021, Proceedings. ed. / Manolis Savva; Ahmed Sabbir Arif. Canadian Information Processing Society, 2021. (Proceedings - Graphics Interface; Vol. 2021-May).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - BayesGaze

T2 - Graphics Interface 2021

AU - Li, Zhi

AU - Zhao, Maozheng

AU - Wang, Yifan

AU - Rashidian, Sina

AU - Baig, Furqan

AU - Liu, Rui

AU - Liu, Wanyu

AU - Beaudouin-Lafon, Michel

AU - Ellison, Brooke

AU - Wang, Fusheng

AU - Ramakrishnan, I. V.

AU - Bi, Xiaojun

PY - 2021

Y1 - 2021

N2 - Selecting targets accurately and quickly with eye-gaze input remains an open research question. In this paper, we introduce BayesGaze, a Bayesian approach of determining the selected target given an eye-gaze trajectory. This approach views each sampling point in an eye-gaze trajectory as a signal for selecting a target. It then uses the Bayes' theorem to calculate the posterior probability of selecting a target given a sampling point, and accumulates the posterior probabilities weighted by sampling interval to determine the selected target. The selection results are fed back to update the prior distribution of targets, which is modeled by a categorical distribution. Our investigation shows that BayesGaze improves target selection accuracy and speed over a dwell-based selection method, and the Center of Gravity Mapping (CM) method. Our research shows that both accumulating posterior and incorporating the prior are effective in improving the performance of eye-gaze based target selection.

AB - Selecting targets accurately and quickly with eye-gaze input remains an open research question. In this paper, we introduce BayesGaze, a Bayesian approach of determining the selected target given an eye-gaze trajectory. This approach views each sampling point in an eye-gaze trajectory as a signal for selecting a target. It then uses the Bayes' theorem to calculate the posterior probability of selecting a target given a sampling point, and accumulates the posterior probabilities weighted by sampling interval to determine the selected target. The selection results are fed back to update the prior distribution of targets, which is modeled by a categorical distribution. Our investigation shows that BayesGaze improves target selection accuracy and speed over a dwell-based selection method, and the Center of Gravity Mapping (CM) method. Our research shows that both accumulating posterior and incorporating the prior are effective in improving the performance of eye-gaze based target selection.

KW - HCI design and evaluation methods

KW - Human computer interaction (HCI)

KW - Human computer interaction (HCI); Human-centered computing

KW - Human-centered computing

KW - Interaction techniques; Human-centered computing

KW - User studies

UR - https://www.scopus.com/pages/publications/85111655254

U2 - 10.20380/GI2021.35

DO - 10.20380/GI2021.35

M3 - Conference contribution

AN - SCOPUS:85111655254

T3 - Proceedings - Graphics Interface

BT - Graphics Interface 2021, Proceedings

A2 - Savva, Manolis

A2 - Arif, Ahmed Sabbir

PB - Canadian Information Processing Society

Y2 - 27 May 2021 through 28 May 2021

ER -

BayesGaze: A bayesian approach to eye-gaze based target selection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this