Automatic Initiation of the Periorbital Signal Extraction in Thermal Imagery
AVSS '09 Proceedings of the 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance
BlinkWrite2: an improved text entry method using eye blinks
Proceedings of the 2010 Symposium on Eye-Tracking Research & Applications
Computer Vision and Image Understanding
Towards hands-free interfaces based on real-time robust facial gesture recognition
AMDO'06 Proceedings of the 4th international conference on Articulated Motion and Deformable Objects
| Hi-index | 0.01 |
Directly connected to the brain, the eyes are the last part of our body we lose control of. For some persons, such as those suffered from a brain-stem stroke, the eyes provide the only means of communication with the world. The eye blinks for such persons are used to make their lexicon and the goal of many rehabilitation centers worldwide is to build tools that would allow automatic detection of the eye blink based lexicon. The tools designed so far are very cumbersome and still do not show the desired performance. At the same time, recent advances in computer hardware and computer vision, in particular, in motion and change detection, offered practitioners a new way for detecting blinks based on video observations of the person's face. This paper overviews different techniques to the problem and describes a vision-based system which is presently being tested in one of the rehabilitation centres. We show how to reliably detect a two-eye blink with a help of an off-the-shelf web-camera and present an approach to the detection a single-eye blink (wink) - this type of blinks is much harder to detect due the lack of spacial constrains, it is however the only type of movement some patients can exhibit.