Gaze stabilization in active vision - I. Vergence error extraction

Michael M. Marefat, Liwei Wu, Christopher C. Yang

Research output: Contribution to journalArticlepeer-review

9 Scopus citations

Abstract

This paper discusses the problems associated with purposeful gazing and fixation of attention for systems with active vision. To help bridge the gap between our current standpoint in active gazing and where advantages that holding gaze brings, we propose solution to the important problems involved in gaze stabilization, i.e. vergence error extraction, and vergence servo control. In the first part, we focus on vergence error extraction. In the next paper, "Gaze Stabilization in Active Vision II: Multi-rate Vergence Control", we shall focus on vergence control. Binocular gazing is realized by decreasing the disparity which represents the vergence error. In order to obtain the disparity for extraction of vergence error, a phase-based approach that efficiently estimates vergence disparity is developed. Experiments are conducted to investigate the effects of window size, segmentation techniques, and illumination.

Original languageEnglish (US)
Pages (from-to)1829-1842
Number of pages14
JournalPattern Recognition
Volume30
Issue number11
DOIs
StatePublished - Nov 1997

Keywords

  • Active vision
  • Gaze stabilization
  • Purposeful gazing
  • Vergence
  • Vision-based dynamic control

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Gaze stabilization in active vision - I. Vergence error extraction'. Together they form a unique fingerprint.

Cite this