A gaze detection system with real-time properties is described. In contrast to most systems available today it is fully passive and only relies on the images taken by a CCD camera without special illumination.To avoid the computational intensive algorithms usually associated with Hough transforms for ellipses (e.g. [Bal81], [TMT8]) a two-stage approach is presented, in which the Hough transform yields a segmented edge image used for a two-dimensional fitting algorithm. The algorithm has been implemented; experimental results are reported.
CITATION STYLE
Klingspohr, H., Block, T., & Grigat, R. R. (1997). A passive real-time gaze estimation system a passive real-time gaze estimation system for human-machine interfaces. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1296, pp. 718–725). Springer Verlag. https://doi.org/10.1007/3-540-63460-6_183
Mendeley helps you to discover research relevant for your work.