Human Motion Reconstruction Based on Multi-view Video

In this paper, the key technologies of human motion reconstruction based on multi-view video are discussed. A more robust motion reconstruction method with markers is presented, which includes camera calibration, markers extraction, markers tracking and 3D reconstruction. In the procedure of markers tracking, we propose a novel multi-view tracking algorithm. This algorithm is made up of two steps, one is binocular stereo vision tracking which combines the Extended Kalman Filter with the smooth of markers´ trajectory, and another is multi-view data fusion. The effectiveness of this method is verified by experiments.