A motion estimation method for incipient facial paralysis based on GAOF

The motion estimation of facial paralysis is a key technique for the diagnosis and evaluation of incipient facial paralysis. In this paper, a novel method is developed to implement the motion estimation of incipient facial paralysis based on GAOF (Optical Flow Method based on Genetic Algorithm). Using this method, the motion vector distribution, features and dynamic pattern were assessed from the dynamic facial image sequences, allowing diagnosis of the site of incipient facial paralysis and assess progression or recovery profiles of patients. Experimental results of some normal and abnormal subjects are reported verifying that this method is effective in the determination of the sites of incipient facial paralysis and quantitative evaluation of recovery process. Compared with other methods, this approach has more advantages such as objectivity, accuracy and convenience for dynamic patterns.