Modeling a 3-D non-rigid virtual object by motion capture and its appearance in the real space

A technique is described for creating a 3-D non-rigid virtual object and projecting it into real space. Real human walking motion is reconstructed employing a developed motion capture system and the obtained 3-D human model is projected into a real space through a see-through type head mounted display. The realized environment is referred to as a mixed reality space. The developed system not only provides smooth motion of the virtual human, but also solves the occlusion problem among the real objects in front of the observer and the projected virtual human. The performed experiment gives satisfactory results.

[1]  Takeo Kanade,et al.  Shape and motion from image streams under orthography: a factorization method , 1992, International Journal of Computer Vision.

[2]  Joo Kooi Tan,et al.  Extracting 3-D motions of individuals at work by uncalibrated multiple video cameras , 1999, IEEE SMC'99 Conference Proceedings. 1999 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.99CH37028).

[3]  Joo Kooi Tan,et al.  An Efficient Technique for Motion Recovery Based on Multiple Views , 1998, MVA.