Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering

Nina Flad; Jonas C. Ditz; Albrecht Schmidt; Heinrich Bulthoff; Lewis L. Chuang

doi:10.1109/ETVIS.2016.7851156

Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering

Nina Flad, Jonas C. Ditz, Albrecht Schmidt, Heinrich Bulthoff, Lewis L. Chuang

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

Abstract

Unrestricted gaze tracking that allows for head and body movements can enable us to understand interactive gaze behavior with large-scale visualizations. Approaches that support this, by simultaneously recording eye-And user-movements, can either be based on geometric or data-driven regression models. A data-driven approach can be implemented more flexibly but its performance can suffer with poor quality training data. In this paper, we introduce a pre-processing procedure to remove training data for periods when the gaze is not fixating the presented target stimuli. Our procedure is based on a velocity-based filter for rapid eye-movements (i.e., saccades). Our results show that this additional procedure improved the accuracy of our unrestricted gaze-Tracking model by as much as 56 %. Future improvements to data-driven approaches for unrestricted gaze-Tracking are proposed, in order to allow for more complex dynamic visualizations.

Original language	English
Title of host publication	Proceedings of the 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-5
Number of pages	5
ISBN (Electronic)	9781509047314
DOIs	https://doi.org/10.1109/ETVIS.2016.7851156
Publication status	Published - 2017 Feb 10
Externally published	Yes
Event	2nd Workshop on Eye Tracking and Visualization, ETVIS 2016 - Baltimore, United States Duration: 2016 Oct 23 → …

Other

Other	2nd Workshop on Eye Tracking and Visualization, ETVIS 2016
Country/Territory	United States
City	Baltimore
Period	16/10/23 → …

Keywords

J.2 [computer applications]: physical sciences and engineering-engineering
J.4 [computer applications]: social and behavioral sciences-psychology

ASJC Scopus subject areas

Human-Computer Interaction
Media Technology

Access to Document

10.1109/ETVIS.2016.7851156

Cite this

Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering. / Flad, Nina; Ditz, Jonas C.; Schmidt, Albrecht et al.
Proceedings of the 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. p. 1-5 7851156.

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

Flad, N, Ditz, JC, Schmidt, A, Bulthoff, H & Chuang, LL 2017, Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering. in Proceedings of the 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016., 7851156, Institute of Electrical and Electronics Engineers Inc., pp. 1-5, 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016, Baltimore, United States, 16/10/23. https://doi.org/10.1109/ETVIS.2016.7851156

@inproceedings{2a784d06aa6b40929e7fc1355c71bb5e,

title = "Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering",

abstract = "Unrestricted gaze tracking that allows for head and body movements can enable us to understand interactive gaze behavior with large-scale visualizations. Approaches that support this, by simultaneously recording eye-And user-movements, can either be based on geometric or data-driven regression models. A data-driven approach can be implemented more flexibly but its performance can suffer with poor quality training data. In this paper, we introduce a pre-processing procedure to remove training data for periods when the gaze is not fixating the presented target stimuli. Our procedure is based on a velocity-based filter for rapid eye-movements (i.e., saccades). Our results show that this additional procedure improved the accuracy of our unrestricted gaze-Tracking model by as much as 56 %. Future improvements to data-driven approaches for unrestricted gaze-Tracking are proposed, in order to allow for more complex dynamic visualizations.",

keywords = "J.2 [computer applications]: physical sciences and engineering-engineering, J.4 [computer applications]: social and behavioral sciences-psychology",

author = "Nina Flad and Ditz, {Jonas C.} and Albrecht Schmidt and Heinrich Bulthoff and Chuang, {Lewis L.}",

year = "2017",

month = feb,

day = "10",

doi = "10.1109/ETVIS.2016.7851156",

language = "English",

pages = "1--5",

booktitle = "Proceedings of the 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2nd Workshop on Eye Tracking and Visualization, ETVIS 2016 ; Conference date: 23-10-2016",

}

TY - GEN

T1 - Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering

AU - Flad, Nina

AU - Ditz, Jonas C.

AU - Schmidt, Albrecht

AU - Bulthoff, Heinrich

AU - Chuang, Lewis L.

PY - 2017/2/10

Y1 - 2017/2/10

N2 - Unrestricted gaze tracking that allows for head and body movements can enable us to understand interactive gaze behavior with large-scale visualizations. Approaches that support this, by simultaneously recording eye-And user-movements, can either be based on geometric or data-driven regression models. A data-driven approach can be implemented more flexibly but its performance can suffer with poor quality training data. In this paper, we introduce a pre-processing procedure to remove training data for periods when the gaze is not fixating the presented target stimuli. Our procedure is based on a velocity-based filter for rapid eye-movements (i.e., saccades). Our results show that this additional procedure improved the accuracy of our unrestricted gaze-Tracking model by as much as 56 %. Future improvements to data-driven approaches for unrestricted gaze-Tracking are proposed, in order to allow for more complex dynamic visualizations.

AB - Unrestricted gaze tracking that allows for head and body movements can enable us to understand interactive gaze behavior with large-scale visualizations. Approaches that support this, by simultaneously recording eye-And user-movements, can either be based on geometric or data-driven regression models. A data-driven approach can be implemented more flexibly but its performance can suffer with poor quality training data. In this paper, we introduce a pre-processing procedure to remove training data for periods when the gaze is not fixating the presented target stimuli. Our procedure is based on a velocity-based filter for rapid eye-movements (i.e., saccades). Our results show that this additional procedure improved the accuracy of our unrestricted gaze-Tracking model by as much as 56 %. Future improvements to data-driven approaches for unrestricted gaze-Tracking are proposed, in order to allow for more complex dynamic visualizations.

KW - J.2 [computer applications]: physical sciences and engineering-engineering

KW - J.4 [computer applications]: social and behavioral sciences-psychology

UR - http://www.scopus.com/inward/record.url?scp=85016006692&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85016006692&partnerID=8YFLogxK

U2 - 10.1109/ETVIS.2016.7851156

DO - 10.1109/ETVIS.2016.7851156

M3 - Conference contribution

AN - SCOPUS:85016006692

SP - 1

EP - 5

BT - Proceedings of the 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd Workshop on Eye Tracking and Visualization, ETVIS 2016

Y2 - 23 October 2016

ER -

Data-driven approaches to unrestricted gaze-Tracking benefit from saccade filtering

Abstract

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this