| Li-Qun Xu, Dave Machin, and Phil Sheppard. A novel approach to real-time nonintrusive gaze finding. In British Machine Vision Conference, 1998. |
....location on the screen that the user is looking at. In their experiments, 2000 training samples were used to train a multilayer neural network, and the authors reported an accuracy of about 1.5 degrees. Their tracker runs at 15Hz, and allows some head movement. A similar method is documented in [14], which also achieved an accuracy of 1.5 degrees, using 3000 training samples. 3 Using Appearance based Methods for Gaze Estimation Instead of using explicit geometric features like contours and corners, an alternative approach to object pose estimation is to treat an image as a point in a ....
....that appearance based parameter estimation is at least a viable solution to the eye gaze estimation problem. Subject Complete Non peripheral X 0.47853 0.44551 Y 0.43685 0.30534 Z 0.47112 0.4004 Mean 0.4622 0. 3838 (a) System Accuracy Calibration (degrees) Samples Xu Machin Sheppard [14] 1.5 3000 Baluja Pomerleau [1] 1.5 2000 ASL Eyelink II 0.5 9 Appearance based 0.38 252 (b) c) Figure 3. Gaze estimation results. a) Average angular errors for the three subjects, measured in degrees. b) Comparing with existing methods. c) Estimation results for subject X. Crosses are ....
L.-Q. Xu, D. Machin, and P. Sheppard. A novel approach to real-time non-intrusive gaze finding. In British Machine Vision Conference, 1998.
No context found.
Li-Qun Xu, Dave Machin, and Phil Sheppard. A novel approach to real-time nonintrusive gaze finding. In British Machine Vision Conference, 1998.
No context found.
L.-Q. Xu, D. Machin, and P. Sheppard. A novel approach to real-time non-intrusive gaze finding. British machine vision conference, Southampton, UK, pages 428--437, 1998.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC