Eye Gaze Estimation from a Single Image of One Eye

  • Authors:
  • Jian-Gang Wang;Eric Sung;Ronda Venkateswarlu

  • Affiliations:
  • -;-;-

  • Venue:
  • ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we present a novel approach, called the"one-circle"algorithm, for measuring the eye gaze using a monocular image thatzooms in on only one eye of a person. Observing that the iriscontour is a circle, we estimate the normal direction of this iriscircle, considered as the eye gaze, from its elliptical image. Frombasic projective geometry, an ellipse can be back-projected intospace onto two circles of different orientations. However, by usingan anthropometric property of the eyeball, the correct solution canbe disambiguated. This allows us to obtain a higher resolutionimage of the iris with a zoom-in camera and thereby achievinghigher accuracies in the estimation. The robustness of our gazedetermination approach was verified statistically by the extensiveexperiments on synthetic and real image data. The two keycontributions in this paper are that we show the possibility offinding the unique eye gaze direction from a single image of oneeye and that one can obtain better accuracy as a consequence ofthis.