Abstract
The term biological motion is often used by researchers studying the patterns of movement generated by living forms and sport forms. We were proposing the human-emotion discrimination system using biological motions in daily life. In this paper, we study the method of a feature extraction to discriminate emotions of walking style, and this is the focus here. The extracted feature and pattern classifier (k-nearest neighbor method) are evaluated from the discriminated result of the given walking data which identified five types of categories, Anger, Sad, Joy, fear, and Neutral.
