RI | Research | Projects | Temporal Shape-From-Silhouette

Text only version of this site

Temporal Shape-From-Silhouette Heads: Takeo Kanade and Simon Baker Contact: Takeo Kanade (tk@cs.cmu.edu) Mailing address: Carnegie Mellon University Robotics Institute 5000 Forbes Avenue Pittsburgh, PA 15213 Associated center: VASC Associated lab/group: Virtualized RealityTM For more information, see this project's homepage.

This page last updated - November 2006.

Project Description

Although Shape-From-Silhouette (SFS) is a popular 3D reconstruction method, the shape estimated using SFS can be coarse if there are only a few cameras. Better shape estimates or Visual Hulls can be obtained if the number of distinct silhouette images is increased. Instead of increasing the number of physical cameras (the across space approach), in this project multiple silhouette images captured across time are combined to compute a refined Visual Hull. The Temporal Shape-From-Silhouette Algorithm consists of two tasks: Visual Hull Alignment and Visual Hull Refinement. Temporal SFS is first devised for rigid objects and then extended to articulated objects (with rigid parts).

Example Results (Rigid Case):

Example Results (Piecewise Rigid/Articulated Case):

Personnel [Past Members]

	Name	Title	Email Address
	Takeo Kanade	U.A. and Helen Whitaker University Prof., RI/CS	tk@cs.cmu.edu

Publications

Note: This list may not be comprehensive. It contains only those publications in the RI publications database. Entries are listed in reverse chronological order.

• Shape-From-Silhouette Across Time: Part II: Applications to Human Modeling and Markerless Motion Tracking
  K.M. Cheung, S. Baker, and T. Kanade
  International Journal of Computer Vision, Vol. 63, No. 3, August, 2005, pp. 225 - 245.
  Download: pdf [2755 KB] copyrighted

• Shape-From-Silhouette Across Time Part I: Theory and Algorithms
  K.M. Cheung, S. Baker, and T. Kanade
  International Journal of Computer Vision, Vol. 62, No. 3, May, 2005, pp. 221 - 247. [Abstract]
  Download: pdf [2047 KB] copyrighted

• Markerless Human Motion Transfer
  K.M. Cheung, S. Baker, J.K. Hodgins, and T. Kanade
  Proceedings of the 2nd International Symposium on 3D Data Processing, Visualization and Transmission, September, 2004. [Abstract]
  Download: pdf [1216 KB] copyrighted

• Markerless Human Motion Transfer
  K.M. Cheung, S. Baker, J.K. Hodgins, and T. Kanade
  In Proceedings of the SIGGRAPH 2004 Conference on Sketches & Applications, ACM Press, August, 2004.
  Download: pdf [423 KB] copyrighted

• Visual Hull Construction, Alignment and Refinement for Human Kinematic Modeling, Motion Tracking and Rendering
  K.M. Cheung
  doctoral dissertation, tech. report CMU-RI-TR-03-44, Robotics Institute, Carnegie Mellon University, October, 2003. [Abstract]
  Download: pdf [8142 KB] copyrighted

• Shape-From-Silhouette of Articulated Objects and its Use for Human Body Kinematics Estimation and Motion Capture
  K.M. Cheung, S. Baker, and T. Kanade
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June, 2003. [Abstract]
  Download: pdf [956 KB] copyrighted

• Visual Hull Alignment and Refinement Across Time: A 3D Reconstruction Algorithm Combining Shape-From-Silhouette with Stereo
  K.M. Cheung, S. Baker, and T. Kanade
  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, June, 2003. [Abstract]
  Download: pdf [931 KB] copyrighted