A Low-Dimensional Radial Silhouette-Based Feature for Fast Human Action Recognition Fusing Multiple Views

Table 4

Comparison of recognition rates and speeds obtained on the MuHAVi-14 dataset with other state-of-the-art approaches.

Approach	LOSO	LOAO	FPS
Singh et al. [8]	82.4%	61.8%	N/A
Eweiwi et al. [17]	91.9%	77.9%	N/A
Cheema et al. [15]	86.0%	73.5%	56
Chaaraoui et al. [5]	91.2%	82.4%	72
Chaaraoui et al. [6]	94.1%	86.8%	51
Our approach	95.6%	88.2%	93