Master Thesis Proposal
“Real-time Human Motion Tracking with Stereo Camera System" 

Motion capture of human movements is prerequisite for realizing skill transfer from humans to robots. Although commercial motion capture systems are available for human motion capturing,these systems are expensive and requires a large set up in the environment. Moreover, human subjects have to wear optical markers or additional sensors on their body. Motion capture of human movements by using robot’s onboard camera system is important for realizing intuitive human-robot interfaces. New technology for onboard camera systems is required to replace the distributed cameras commonly used for human motion capture. In this thesis a method for three dimensional human motion capturing using robot’s onboard stereo camera system should be developed.

 

Detailed research issues are
• Review of state of the art in markerless human motion capturing
• Probabilistic modelling of articulated human motions
• 3D motion capturing algorithm using a stereo camera
• Real-time computation
• Handling self-occlusion
• Evaluation of implemented algorithms in real experiments
• Integration with NAO's motion imitation algorithm (optional)
• Handling fast movements (optional)
• Handling dynamic background (optional)

 

Related works

Dongheui Lee and Yoshihiko Nakamura , Motion Capturing from Monocular Vision by Statistical Inference Based on Motion Database: Vector Field Approach , 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'07) , 2007 , p. 617-623 .

Azad, P., Ude, A., Asfour, T., Dillmann, R., Stereo-based Markerless Human Motion Capture for Humanoid Robot Systems, Proceedings of IEEE International Conference on Robotics and Automation 2007, p.3951-3956.

 

Student
Qixun Liang

 
Supervisor
Prof. Dongheui Lee