Perceptually optimised coding of open signed video based on gaze prediction

A gaze location prediction model is presented for open signed broadcast video based on actual eye tracking data with fluent British sign language users. The model is used for formulating a variable quality coding approach of such video which is shown to be capable of offering up to 30% reduction in bit rate with little or no discernible effect on the understanding or viewing experience, as suggested by the results of a subjective evaluation by sign language viewers.