Tridimensional Pose Estimation of Person Head

PEREZ ELISA ; SORIA CARLOS; NASISI OSCAR; MUT VICENTE

San Juan

Congreso; XVICongreso Argentino de Bioingeniería, V Jornadas de Ingeniería Clínica; 2007

Universidad Nacional de San Juan- Facultad de Ingeniería

In this work, we present a method for estimating 3-D motion parameters; this method provides an alternative way for 3D head pose estimation from image sequence in the current computer vision literature. This method is robust over extended sequences and large head motions and accurately extracts the orientation angles of head from a single view. Experimental results show that this tracking system works well for the development of a human-computer interface for people that had severe motor disabilities.