The goal of multimodal human-computer interaction is the synergetic use
of several (and ultimately, all) of the sensing and
expressing abilities that humans have (Jaimes and Sebe 2005). Analysing
the video channel is very important in achieving
natural and human-like human-computer communication. The face carries
many characteristics of the interlocutor: gender,
age, race, etc. and it also can convey a lot of additional information
during the conversation: facial expressions,
gestures, grimaces, gaze, head-shake. Research within the domain of face
analysis has been conducted for many years.
In order to resolve the research questions listed above, we give a
general survey of this topics.
|