Human Model Adaptation for Multiview Markerless Motion Capture

Figure 5

Shape estimation results based eight cameras: (a) fitting result of the first frame, (b) fitting result of the 25th frame.

(a)

(b)