Expression Recognition by Using Facial and Vocal Expressions

Abstract

Human behaviour may be monitored by analysing facial expressions and vocal expressions. Hence an automatic technique which combines both these features will give a more accurate overall estimation of expression. In this work we propose a new method which is uses facial and vocal features to estimate the expression of the subject. Facial expressions are analysed by extracting important facial features and then clustering the movement of these features. In parallel the voice is processed by using considering sudden changes in amplitude and frequency in order to recognize the expression. Finally a weighted sum rule is used to combine the decisions obtained by facial and vocal expression recognition. The proposed technique is tested on an ongoing set of real data monitored by a psychologist.

Dominant and Complementary Multi-Emotional Facial Expression Recognition Using C-Support Vector Classification

“…Automatic facial expression recognition is widely used in many applications such as assistive robotics, human emotion analysis, and human-machine interfaces [7], [8]. As a result, recognition of facial expressions has become an important research topic in human-computer interaction, as well as in the fields of machine learning, image processing, human cognition, and pattern recognition.…”