Skeleton-Based Action Recognition Based on Distance Vector and Multihigh View Adaptive Networks

Table 2

Results on the PKU-MMD datasets.

Methods	X-Sub (%)	X-View (%)
STA-LSTM [29]	44.4	13.1
JCRRNN [32]	32.5	53.3
Skeleton boxes [33]	54.8	94.2
Li et al. [34]	90.4	93.7
HCN [6]	92.6	94.2
Ours	92.7	94.4