Circle-based Eye Center Localization (CECL)

Yustinus Soelistio, Eric Postma, Alfons Maes

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Abstract

The ability to automatically detect eye center locations in video images allows for estimating gaze direction. This, in turn, facilitates the study of human-computer interaction and behavioral analyses of social interactions. We propose an improved eye center localization method based on the Hough transform, called Circle-based Eye Center Localization (CECL) that is simple, robust, and achieves accuracy at a par with typically more complex state-of-the-art methods. The CECL method relies on color and shape cues that distinguish the iris from other facial structures. The circle enclosing the iris is localized by means of the Hough transform and the center of the iris is determined using the intensity level within the detected circle. The accuracy of the CECL method is demonstrated through a comparison with 15 state-of-the-art eye center localization methods against five error thresholds, as reported in the literature. The CECL method achieved an accuracy of 80.8% to 99.4% and ranked first for 2 of the 5 thresholds. It is concluded that the CECL method offers an attractive alternative to existing methods for automatic eye center localization.
Original languageEnglish
Title of host publicationMVA 2015 IAPR International Conference on Machine Vision Applications
PublisherIEEE
Pages349-352
Number of pages4
DOIs
Publication statusPublished - 18 May 2015

Fingerprint

Hough transforms
Human computer interaction
Color

Cite this

Soelistio, Y., Postma, E., & Maes, A. (2015). Circle-based Eye Center Localization (CECL). In MVA 2015 IAPR International Conference on Machine Vision Applications (pp. 349-352). [10-30] IEEE. https://doi.org/10.1109/MVA.2015.7153202
Soelistio, Yustinus ; Postma, Eric ; Maes, Alfons. / Circle-based Eye Center Localization (CECL). MVA 2015 IAPR International Conference on Machine Vision Applications. IEEE, 2015. pp. 349-352
@inproceedings{fce91598f56144c48ad400e76cf35384,
title = "Circle-based Eye Center Localization (CECL)",
abstract = "The ability to automatically detect eye center locations in video images allows for estimating gaze direction. This, in turn, facilitates the study of human-computer interaction and behavioral analyses of social interactions. We propose an improved eye center localization method based on the Hough transform, called Circle-based Eye Center Localization (CECL) that is simple, robust, and achieves accuracy at a par with typically more complex state-of-the-art methods. The CECL method relies on color and shape cues that distinguish the iris from other facial structures. The circle enclosing the iris is localized by means of the Hough transform and the center of the iris is determined using the intensity level within the detected circle. The accuracy of the CECL method is demonstrated through a comparison with 15 state-of-the-art eye center localization methods against five error thresholds, as reported in the literature. The CECL method achieved an accuracy of 80.8{\%} to 99.4{\%} and ranked first for 2 of the 5 thresholds. It is concluded that the CECL method offers an attractive alternative to existing methods for automatic eye center localization.",
author = "Yustinus Soelistio and Eric Postma and Alfons Maes",
year = "2015",
month = "5",
day = "18",
doi = "10.1109/MVA.2015.7153202",
language = "English",
pages = "349--352",
booktitle = "MVA 2015 IAPR International Conference on Machine Vision Applications",
publisher = "IEEE",

}

Soelistio, Y, Postma, E & Maes, A 2015, Circle-based Eye Center Localization (CECL). in MVA 2015 IAPR International Conference on Machine Vision Applications., 10-30, IEEE, pp. 349-352. https://doi.org/10.1109/MVA.2015.7153202

Circle-based Eye Center Localization (CECL). / Soelistio, Yustinus; Postma, Eric; Maes, Alfons.

MVA 2015 IAPR International Conference on Machine Vision Applications. IEEE, 2015. p. 349-352 10-30.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

TY - GEN

T1 - Circle-based Eye Center Localization (CECL)

AU - Soelistio, Yustinus

AU - Postma, Eric

AU - Maes, Alfons

PY - 2015/5/18

Y1 - 2015/5/18

N2 - The ability to automatically detect eye center locations in video images allows for estimating gaze direction. This, in turn, facilitates the study of human-computer interaction and behavioral analyses of social interactions. We propose an improved eye center localization method based on the Hough transform, called Circle-based Eye Center Localization (CECL) that is simple, robust, and achieves accuracy at a par with typically more complex state-of-the-art methods. The CECL method relies on color and shape cues that distinguish the iris from other facial structures. The circle enclosing the iris is localized by means of the Hough transform and the center of the iris is determined using the intensity level within the detected circle. The accuracy of the CECL method is demonstrated through a comparison with 15 state-of-the-art eye center localization methods against five error thresholds, as reported in the literature. The CECL method achieved an accuracy of 80.8% to 99.4% and ranked first for 2 of the 5 thresholds. It is concluded that the CECL method offers an attractive alternative to existing methods for automatic eye center localization.

AB - The ability to automatically detect eye center locations in video images allows for estimating gaze direction. This, in turn, facilitates the study of human-computer interaction and behavioral analyses of social interactions. We propose an improved eye center localization method based on the Hough transform, called Circle-based Eye Center Localization (CECL) that is simple, robust, and achieves accuracy at a par with typically more complex state-of-the-art methods. The CECL method relies on color and shape cues that distinguish the iris from other facial structures. The circle enclosing the iris is localized by means of the Hough transform and the center of the iris is determined using the intensity level within the detected circle. The accuracy of the CECL method is demonstrated through a comparison with 15 state-of-the-art eye center localization methods against five error thresholds, as reported in the literature. The CECL method achieved an accuracy of 80.8% to 99.4% and ranked first for 2 of the 5 thresholds. It is concluded that the CECL method offers an attractive alternative to existing methods for automatic eye center localization.

U2 - 10.1109/MVA.2015.7153202

DO - 10.1109/MVA.2015.7153202

M3 - Conference contribution

SP - 349

EP - 352

BT - MVA 2015 IAPR International Conference on Machine Vision Applications

PB - IEEE

ER -

Soelistio Y, Postma E, Maes A. Circle-based Eye Center Localization (CECL). In MVA 2015 IAPR International Conference on Machine Vision Applications. IEEE. 2015. p. 349-352. 10-30 https://doi.org/10.1109/MVA.2015.7153202