CALL FOR PAPERS
Computer Vision and Image Understanding
Special Issue on Eye Detection and Tracking
Guest Editors:
Qiang Ji, Rensselaer Polytechnic Institute, qji@ecse.rpi.edu
Harry Wechsler, George Mason University, wechsler@cs.gmu.edu
Andrew Duchowski, Clemson University, andrewd@vr.clemson.edu
Myron Flickner, IBM Almaden Research, flick@almaden.ibm.com
AIMS AND SCOPE
As one of the most salient features of human face, eyes play an important
role in interpreting and understanding a person's desires, needs, and
emotional
states. Robust non-intrusive eye detection and tracking is, therefore,
crucial for human computer interaction, attentive user interfaces, and
understanding human affective states. In addition, the unique geometric,
photometric, and motion characteristics of the eyes also provide important
visual cues for face detection, face recognition, and for facial expression
understanding.
There has been much work in eye detection and tracking. The existing work
can be broadly classified into two categories: traditional image based
passive approaches and the active IR based approaches. The former approaches
can be further divided into appearance-based, model-based, feature-based,
and motion-based methods. The latter approach exploits the spectral properties
of pupil under near IR illumination. Eye tracking is accomplished by tracking
the bright (dark) pupils. Recently, there are efforts in combining the
passive approach with the active approach to produce more robust eye tracking.
Despite these efforts, robust, accurate, and non-intrusive eye detection and
tracking remains largely an unsolved issue. The challenges result from eye
closure, eye occlusion, variability in scale and face orientation, and
different lighting conditions. The many commercial eye trackers tend to be
intrusive and restrictive. Furthermore, they often require a cumbersome
calibration process. On the other hand, the existing non-intrusive eye
tracking techniques lack robustness and accuracy which limit their use in
practice.
The special issue solicits original research that focuses on the following
aspects of eye detection and gaze estimation
* non-intrusive eye detection and tracking
* non-intrusive eye gaze estimation requiring minimal or no user calibration
* eye gestures and activities detection and characterization
* applications of eye detection and tracking techniques
* comprehensive review/survey of the existing technologies in eye and gaze
detection and tracking
SUBMISSION GUIDELINES
Only original, high-quality papers -- in-line with the CVIU guidelines --
will be considered for publication in this special issue. Prospective authors
concerned about their paper falling under the scope of this special issue,
should send an abstract to the guest editors for a preliminary evaluation
prior to the submission of the full paper.
All papers should be submitted electronically via our http server. Files
should be in PDF or PS format. Authors should also submit a cover letter
in plain text with the following information: title of the submitted article,
the name of the file that has been submitted, all authors full names, and,
the corresponding author's mailing address, day time phone number and e-mail.
Please, send all cover letters via e-mail to Prof. Qiang Ji at
qji@ecse.rpi.edu
The URL for our http server is http://cviu.ecse.rpi.edu/cviu/login.php
Follow the instructions there to submit your paper.
If you use LaTeX, please use elsart style files to format your manuscript.
The latex styles can be downloaded from http://www.elsevier.com/locate/latex
IMPORTANT DATES:
* Electronic submission of full manuscripts: October 15,
2003
* Notification to authors: February 15, 2004
* Submission of revised manuscripts: April 15, 2004
* Final decision on accepted papers: May 15, 2004
* Publication of special issue: Third quarter 2004