3D Computation for Motion Capture

3D computation in the motion capture project is similar to computation in a digitizing project. The only additional issues are:

1. how to effectively compute camera parameters in multiple frames, and;
2. how to compute the x, y, z coordinates in multiple frames.

When all the videos are taken by fixed cameras, we only need to compute camera parameters in the current frame. (See 3D computation for videos taken by fixed cameras.)

When dynamic cameras are used, we compute the camera parameters and x, y, z coordinates at the same time. (See 3D computation for videos taken by dynamic cameras.)