The recognition of facial expressions from video frames

Abstract

Several researchers have attempted automatic recognition and tracking of facial expressions in video frames of neutral and expressive faces. However, it is very hard to detect a neutral face in practical applications because the distinction between facial movement and head movement is difficult to make. In order to solve this distinction problem, the initialization of the head position using the central point of each eye and the tip of the nose is first made. Facial movements are then extracted. Based on the detected neutral faces, the changing processes of facial expressions can be recognized using the recognition algorithm for facial expressions which is described based on the fuzzy expert system. Computer simulation using 48 expressive facial videos shows more than 77% accuracy on average.