Citation link:
Authors:
Муратовски, Красимир; Muratovski, Krasimir; Бумбаров, Огнян Любенов; Boumbarov, Ognian Lyubenov
Title:
Проследяване на лица с адаптивен Mean-Shift алгоритъм и филтър на Калман
Other Titles:
Face tracking using adaptive Mean-Shift algorithm and Kalman filter
Date of Issue:
2008
Is Part of:
Proceedings of the Technical University – Sofia, 58(1), 2008, pp. 118-127
Publisher:
Технически университет - София
Identifiers:
1311-0829 [issn]
Type:
Article
Language:
eng
Abstract:
This paper presents a novel approach to face detection and tracking in video sequences. Probability distribution of the color for the human skin is used as major facial feature. A 2D histogram built from the color components of the image in YCrCb color space is used as representation of the model of the human skin. Facetracking involves the mean-shift algorithm. We propose a method for adaptation both of the tracking-window and of the color-distribution model in order to increase robustness to illumination changes of the environment. Minimizing of the number of iterations is achieved by using dynamic prediction with Kalman filter.
Description:
Reviewer prof. PhD Alexander Bekjarski. - Formulas, pictures, diagr. - References, pp. 127
