Facial Emotional Expressions of Life-like Character Based on Text Classifier and Fuzzy Logic

A system consists of a text classifier and Fuzzy Inference System (FIS) to build a life-like virtual character capable of expressing emotion from a text input is proposed. The system classifies emotional content of sentence(s) from text input and expresses corresponding emotion by a facial expression. Text input is classified using the text classifier, while facial expression of the life-like character are controlled by FIS utilizing results from the text classifier. A number of text classifier methods are employed and their performances are evaluated using Leave-One-Out Cross Validation. In real world application such as animation movie, the lifelike virtual character of proposed system needs to be animated. As a demonstration, examples of facial expressions with corresponding text input as results from the implementation of our system are shown. The system is able to show facial expressions with admixture blending emotions. This paper also describes animation characteristics of the system using neutral expression as center of facial expression transition from one emotion to another. Emotion transition can be viewed as gradual decrease or increase of emotion intensity from one emotion toward other emotion. Experimental results show that animation of lifelike character expressing emotion transition can be generated automatically using proposed system.