In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass

Shoya Ishimaru, Kai Steven Kunze, Koichi Kise, Jens Weppner, Andreas Dengel, Paul Lukowicz, Andreas Bulling

Research output: Chapter in Book/Report/Conference proceedingConference contribution

41 Citations (Scopus)

Abstract

We demonstrate how information about eye blink frequency and head motion patterns derived from Google Glass sensors can be used to distinguish different types of high level activities. While it is well known that eye blink frequency is correlated with user activity, our aim is to show that (1) eye blink frequency data from an unobtrusive, commercial platform which is not a dedicated eye tracker is good enough to be useful and (2) that adding head motion patterns information significantly improves the recognition rates. The method is evaluated on a data set from an experiment containing five activity classes (reading, talking, watching TV, mathematical problem solving, and sawing) of eight participants showing 67% recognition accuracy for eye blinking only and 82% when extended with head motion patterns.

Original languageEnglish
Title of host publicationProceedings of the 5th Augmented Human International Conference, AH 2014
PublisherAssociation for Computing Machinery
ISBN (Print)9781450327619
DOIs
Publication statusPublished - 2014
Externally publishedYes
Event5th Augmented Human International Conference, AH 2014 - Kobe, Japan
Duration: 2014 Mar 72014 Mar 8

Other

Other5th Augmented Human International Conference, AH 2014
CountryJapan
CityKobe
Period14/3/714/3/8

Fingerprint

Sawing
Glass
Sensors
Experiments

Keywords

  • Activity Recognition
  • Blink Frequency
  • Google Glass
  • Head Mounted Sensor
  • IMU
  • Infrared Proximity Sensor

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Ishimaru, S., Kunze, K. S., Kise, K., Weppner, J., Dengel, A., Lukowicz, P., & Bulling, A. (2014). In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass. In Proceedings of the 5th Augmented Human International Conference, AH 2014 [a15] Association for Computing Machinery. https://doi.org/10.1145/2582051.2582066

In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass. / Ishimaru, Shoya; Kunze, Kai Steven; Kise, Koichi; Weppner, Jens; Dengel, Andreas; Lukowicz, Paul; Bulling, Andreas.

Proceedings of the 5th Augmented Human International Conference, AH 2014. Association for Computing Machinery, 2014. a15.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ishimaru, S, Kunze, KS, Kise, K, Weppner, J, Dengel, A, Lukowicz, P & Bulling, A 2014, In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass. in Proceedings of the 5th Augmented Human International Conference, AH 2014., a15, Association for Computing Machinery, 5th Augmented Human International Conference, AH 2014, Kobe, Japan, 14/3/7. https://doi.org/10.1145/2582051.2582066
Ishimaru S, Kunze KS, Kise K, Weppner J, Dengel A, Lukowicz P et al. In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass. In Proceedings of the 5th Augmented Human International Conference, AH 2014. Association for Computing Machinery. 2014. a15 https://doi.org/10.1145/2582051.2582066
Ishimaru, Shoya ; Kunze, Kai Steven ; Kise, Koichi ; Weppner, Jens ; Dengel, Andreas ; Lukowicz, Paul ; Bulling, Andreas. / In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass. Proceedings of the 5th Augmented Human International Conference, AH 2014. Association for Computing Machinery, 2014.
@inproceedings{9d76c83aeb334940b29019b5624ca501,
title = "In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass",
abstract = "We demonstrate how information about eye blink frequency and head motion patterns derived from Google Glass sensors can be used to distinguish different types of high level activities. While it is well known that eye blink frequency is correlated with user activity, our aim is to show that (1) eye blink frequency data from an unobtrusive, commercial platform which is not a dedicated eye tracker is good enough to be useful and (2) that adding head motion patterns information significantly improves the recognition rates. The method is evaluated on a data set from an experiment containing five activity classes (reading, talking, watching TV, mathematical problem solving, and sawing) of eight participants showing 67{\%} recognition accuracy for eye blinking only and 82{\%} when extended with head motion patterns.",
keywords = "Activity Recognition, Blink Frequency, Google Glass, Head Mounted Sensor, IMU, Infrared Proximity Sensor",
author = "Shoya Ishimaru and Kunze, {Kai Steven} and Koichi Kise and Jens Weppner and Andreas Dengel and Paul Lukowicz and Andreas Bulling",
year = "2014",
doi = "10.1145/2582051.2582066",
language = "English",
isbn = "9781450327619",
booktitle = "Proceedings of the 5th Augmented Human International Conference, AH 2014",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - In the blink of an eye - Combining head motion and eye blink frequency for activity recognition with google glass

AU - Ishimaru, Shoya

AU - Kunze, Kai Steven

AU - Kise, Koichi

AU - Weppner, Jens

AU - Dengel, Andreas

AU - Lukowicz, Paul

AU - Bulling, Andreas

PY - 2014

Y1 - 2014

N2 - We demonstrate how information about eye blink frequency and head motion patterns derived from Google Glass sensors can be used to distinguish different types of high level activities. While it is well known that eye blink frequency is correlated with user activity, our aim is to show that (1) eye blink frequency data from an unobtrusive, commercial platform which is not a dedicated eye tracker is good enough to be useful and (2) that adding head motion patterns information significantly improves the recognition rates. The method is evaluated on a data set from an experiment containing five activity classes (reading, talking, watching TV, mathematical problem solving, and sawing) of eight participants showing 67% recognition accuracy for eye blinking only and 82% when extended with head motion patterns.

AB - We demonstrate how information about eye blink frequency and head motion patterns derived from Google Glass sensors can be used to distinguish different types of high level activities. While it is well known that eye blink frequency is correlated with user activity, our aim is to show that (1) eye blink frequency data from an unobtrusive, commercial platform which is not a dedicated eye tracker is good enough to be useful and (2) that adding head motion patterns information significantly improves the recognition rates. The method is evaluated on a data set from an experiment containing five activity classes (reading, talking, watching TV, mathematical problem solving, and sawing) of eight participants showing 67% recognition accuracy for eye blinking only and 82% when extended with head motion patterns.

KW - Activity Recognition

KW - Blink Frequency

KW - Google Glass

KW - Head Mounted Sensor

KW - IMU

KW - Infrared Proximity Sensor

UR - http://www.scopus.com/inward/record.url?scp=84899796433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899796433&partnerID=8YFLogxK

U2 - 10.1145/2582051.2582066

DO - 10.1145/2582051.2582066

M3 - Conference contribution

AN - SCOPUS:84899796433

SN - 9781450327619

BT - Proceedings of the 5th Augmented Human International Conference, AH 2014

PB - Association for Computing Machinery

ER -