Active Shape Model Based Recognition Of Facial Expression

The tracking and recognition of facial activities from image or video is useful in many applications such as animation and human machine interaction. The facial activities are described in three levels. In the bottom level, the facial components are detected. In the middle level, movements in the facial components can be identified. The top level represents the facial muscle movement and human emotion. Eigenfaces algorithm is used to detect face from the image. Hough Transform is used to extract the features. Naive Bayes Classifier and Active Shape Model are used to classify the features. The DBN is introduced between the Hough Transform and the classifier.