Comparative Study of Methods for Recognizing Human Actions from a Real Video Sequence

摘要

This paper explores the effectiveness of using three image features instead of synthesized human motion data by using the real video sequence. We have compared three algorithms that recognize the observed action generated by an unknown person, who is not included in the database. We tested the 4 methods using 3 single image features with 4 human actors and 5 classes of action. In addition, we proved all the three method are useful for the human action recognition.