Human Kinematic Modeling and Motion Capture - Robotics Institute Carnegie Mellon University
Graphical depiction of the Human Kinematic Modeling and Motion Capture project
Human Kinematic Modeling and Motion Capture
Project Head: Takeo Kanade

We have constructed a system for building 3D kinematic models (shape, body part segmentation, and joint information) of humans. This system is based on our temporal shape-from-silhouette algorithms. An overview of the system is as follows:


An example of the results of applying the system is the three kinematic models illustrated in the following movie.

These kinematic models can then be used to track new motions of the person using a tracking algorithm closely related to our temporal shape-from-silhouette
algorithms. Examples of tracking the person in this way are shown below.

Displaying 6 Publications

2016
Master's Thesis, Tech. Report, CMU-RI-TR-16-22, Robotics Institute, Carnegie Mellon University, May, 2016
2005
Kong Man Cheung, Simon Baker, and Takeo Kanade
Journal Article, International Journal of Computer Vision, Vol. 63, No. 3, pp. 225 - 245, August, 2005
Kong Man Cheung, Simon Baker, and Takeo Kanade
Journal Article, International Journal of Computer Vision, Vol. 62, No. 3, pp. 221 - 247, May, 2005
2004
Conference Paper, Proceedings of 2nd International Symposium on 3D Data Processing, Visualization and Transmission (3DPVT '04), pp. 373 - 378, September, 2004
Conference Paper, Proceedings of SIGGRAPH '04 Conference on Sketches & Applications: Session: Motion, pp. 31, August, 2004
2003
Kong Man Cheung, Simon Baker, and Takeo Kanade
Conference Paper, Proceedings of (CVPR) Computer Vision and Pattern Recognition, pp. 77 - 84, June, 2003

current head

current staff

current contact

past head

  • Simon Baker