|
|
Head pose, eye gaze and facial gestures convey a significant narrative
of the user's state and should control the flow of the human computer
interactions. Head pose and eye gaze jointly reveal the spatial point
or region of interest of the user. Facial expressions and actions
convey the emotions or attentional state of the user.
In this research effort we have developed a technique that permits us
to automatically detect and track the head pose and to derive the six
degrees of freedom parameters of the head motion from a single
camera. It enables us to identify the head pose that will be used in
the next stage of robust facial gesture identification that is independent
of the head pose.
|