Modeling of human motion through motion captured data

Abstract

In this project, types of human motion studied in kinematics field are walking and running. The area of interest of the analyzing is lower part of human body segment. A video camera with medium resolution and frame rate is used as an image device to capture the human motion. The image is processed and analyzed to determine the location of every marker at each sequence of motion. Each marker represents a specific part of human body. The extracted data of location for specified limbs used as a basic data. This basic data of coordinate is transformed intoangle data to be the row data i modeling the motion using non-linear identification. The parametric methods are used in modeling human motion. Finally, the identified models are validated using some model validation techniques.