Contents of this page are obsolete. This page is preserved and stored at this URL just from historical viewpoint. Original URL was http://www.mm.media.kyoto-u.ac.jp/members/kameda/...
Please visit www.kameda-lab.org for recent information. (2002/12/06, kameda@ieee.org)

Publication, Research, Docs and Info, KAMEDA


next up previous
Next: Conclusion Up: Three Dimensional Motion Estimation Previous: Algorithm

Experiment

We applied our method to an image sequence with a male walking. The image size 600 by 360 pixels and the sequence includes 40 frames. is set to 6 degree and is 2,500 pixels. The estimated result is shown in Figure 2. The result motion is displayed 100 cm right to the actual result location for the readers. Here you can see that our method keeps tracking the parts in spite of occlusions. As our method uses a complete human model and can obtain the joint angles from the image sequence, we can show the estimation result from any viewpoint. See Figure 3.

  
Figure 2: Result

  
Figure 3: Result from Another Viewpoint



Yoshinari Kameda
Mon Apr 7 14:56:47 JST 1997