Detecting, modeling and tracking of a short-term mutual awareness activity

It is quite common that multiple human observers are attending to a single point-of-interest. Mutual awareness activity (MAWA) refers to the dynamic of this social phenomena. A preferred way to monitor this social phenomenon is with a camera that captures the human observers. The current work studies the short-term dynamics of a MAWA. A fully unsupervised online method is suggested that can deal with the general case of an uncalibrated camera in a general environment and an unconstrained activity in the scene. This is in contrast to other work on similar problems that inherently assume a known environment or a calibrated camera or a restricted activity in the scene. The method was tested on a short video and robustly detected the MAWA and estimated its related attributes. A correlation between the attributes of the detected MAWA and the environment´s events was used to evaluate the method.