Examination of human factors for wearable line-of-sight detection system

Miho Ogawa, Kota Sampei, Carlos Cortes, Norihisa Miki

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

Abstract

We proposed a wearable line-of-sight detection system that utilizes micro-fabricated transparent optical sensors on eyeglasses. These sensors can detect the reflection of light from the eye, in which the intensity from the white of the eye is stronger than that of the pupil, and can thus deduce the position of the pupil. LOS detection was successfully demonstrated by using the proposed system, but careful calibration was required for each user. Therefore, in the current study, we investigated the dominant factors that affected the LOS detection accuracy. It was experimentally found that the distance between the sensors on the eyeglasses and the pupil was a dominant factor. Thus, we designed a frame that can be adjusted according to this distance, which enabled LOS detection for all subjects.

Original languageEnglish
Title of host publicationISWC 2014 - Proceedings of the 2014 ACM International Symposium on Wearable Computers
PublisherAssociation for Computing Machinery
Pages139-140
Number of pages2
ISBN (Electronic)9781450329699
DOIs
Publication statusPublished - 2014 Sep 13
Event18th ACM International Symposium on Wearable Computers, ISWC 2014 - Seattle, United States
Duration: 2014 Sep 132014 Sep 17

Publication series

NameProceedings - International Symposium on Wearable Computers, ISWC
ISSN (Print)1550-4816

Conference

Conference18th ACM International Symposium on Wearable Computers, ISWC 2014
Country/TerritoryUnited States
CitySeattle
Period14/9/1314/9/17

Keywords

  • line-of-sight
  • microelectromechanical systems
  • wearable

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Examination of human factors for wearable line-of-sight detection system'. Together they form a unique fingerprint.

Cite this