Visual Tracking Based on an Improved Online Multiple Instance Learning Algorithm

Table 3

The average computational cost (s) for different algorithms conducted on 6 video clips.

Video clip	MIL	WMIL	Significance-coefficients MIL	IMIL
David indoor	1.122	0.109	1.021	0.084
Occluded face 2	1.601	0.104	1.108	0.087
Coke can	1.089	0.104	0.81	0.104
Cliff bar	1.092	0.098	0.962	0.079
Tiger 2	1.092	0.103	1.096	0.094
Coupon book	1.164	0.103	0.978	0.086