A Lightweight Hierarchical Model with Frame-Level Joints Adaptive Graph Convolution for Skeleton-Based Action Recognition

Table 5

Performance comparison of NTU60.

	Methods	Year	CS	CV
CNN	HCN [8]	2018	86.5	91.1
	View-invariant CNN [9]	2019	86.7	91.8
RNN	Ind-RNN [13]	2018	81.8	88.0
	2s ARRN-LSTM [14]	2019	81.8	89.4
	AGC-LSTM [36]	2019	89.2	95.0
GCN	ST-GCN [19]	2018	81.5	88.3
	2s-AGCN [20]	2019	88.5	95.1
	SGN [26]	2020	89.0	94.5
Ours	FLAGCN	−	89.4	94.8