Research on Discriminative Skeleton-Based Action Recognition in Spatiotemporal Fusion and Human-Robot Interaction

Figure 1

Local posture motion-based temporal attention (LMP-TAM) module.