Mapping of Facial Elements for Emotion Analysis

The recognition of emotions from others´ face is an universal and fundamental skill for social interaction. Many researchers argue that there is a set of basic emotions which were preserved during evaluative process because they allow the adaption of the organisms behaviour to distinct daily situations. This paper is about emotion analysis and the identification of facial elements relevant for this task. The proposed approach is based on psychological research. Different facial representations are proposed to represent the geometrical characteristics of the human face and their performance is evaluated using Machine Learning techniques. Also, an investigation of the facial elements related to the identification of the six basic emotions plus the neutral state is presented.