Figure 2 From Action Recognition With Spatio Temporal Visual Attention On Skeleton Image