Title :
Face location and eye character detection based on monocular vision gaze tracking system
Author :
Wang, Fenhua ; Chen, Xiao ; Chi, Jiannan ; Wang, Zhiliang
Author_Institution :
Minist. of Educ. Key Lab., Univ. of Sci. & Technol., Beijing, China
Abstract :
In the paper an face location and eye detection method is proposed based on active infrared source system, which is to solve the problems of pupil threshold segmentation according to from coarse and fine strategy. This method use bright pupil and dark pupil difference scheme. Firstly, we collect a lot of dark images to train classifiers make use of Haar features, boosting them to be a strong classifiers to face location; then potential region of pupil is found based on the knowledge about facial structure; Finally, human eyes location can be precise found.
Keywords :
Haar transforms; computer vision; Haar features; active infrared source system; bright pupil difference scheme; dark images; dark pupil difference scheme; eye character detection; eye detection method; face location; human eyes location; monocular vision gaze tracking system; Classification algorithms; Face; Face detection; Feature extraction; Image edge detection; Image segmentation; cascade classifiers; dark and bright pupil image; eye location; face location; gaze tracking;
Conference_Titel :
Network Infrastructure and Digital Content, 2010 2nd IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-6851-5
DOI :
10.1109/ICNIDC.2010.5657948