Gesture Recognition and Motion Capture
Gesture Recognition and Motion Capture as a Precursor for Behavioral Biometrics (2006)
MUNS Macquarie University New Staff Research Grant
Investigators: Kavakli, M
This project investigates motion tracking and behavioural modelling as a precursor for the design and implementation of a behavioural biometrics system. The main goals of this project is first, to explore how to produce realistic representations of avatars simulating natural motions based on behavioural protocols, investigating social cues, and then, to produce personality and emotion models in order to use them in personal identification as a part of behavioural biometrics.
Biometrics can be defined as automated methods of identifying a person, or verifying the identity of a person, based on physiological or behavioural characteristics. It is the most secure and convenient authentication tool. The outcomes are behavioural protocols including the personality and emotion models as a precursor for behavioural biometrics. Our hypothesis is that any intelligent agent must be able to detect and process the social cues to be able to operate in a social context in a virtual environment.
While presentation of physical and social cues is a topic of animation, their recognition is an active research area in biometrics. We will specifically focus on the representation of the manner of walking and gait recognition, since gait information is one of the most appropriate methods in behavioural biometrics. The major advantage of gait recognition is its high acceptance rate by the public, since it is based on natural walking. In this project, our aim is to investigate how to improve the accuracy rate in personal identification by addressing the problems in behavioural biometrics.
Although a few studies attempted gait and foot print recognition as an authentication tool, none of these has explored behavioural characteristics in a social context. We will use a motion capture approach based on magnetic sensors that allows the identification of full body movement. Taking advantage of the magnetic sensor technology, we will explore ways, interactively, realistically, and efficiently create and animate virtual humans. We will use Softimage for avatar construction and a number of Virtual Reality tools, such as a Motion Capture suit, Motion Builder Software for tracking motions, and a SpacePad to capture human motions for animations. We will conduct a series of experiments exporting the virtual humans we create to a Game Engine.
We have been trained on how to track motions using a Motion Capture suit, and how to import and manipulate motion capture data in Motion Builder. We also obtained Ethics approval for testing. In addition to a Digital Artist working as a Research Assistant in this project, we have a VR programmer. The project also proceeds parallel to an Honours thesis.