Multiscale hypergraph convolutional network for skeleton-based action recognition

Table 1. Ablation study of HCB and HMB
View table
View in Article
Table 1. Ablation study of HCB and HMB
方法 CV精度/%
Baseline 95.3
Baseline+ ${\varepsilon }_{10}$ 96.0
Baseline+ ${\varepsilon }_{10}$+ ${\varepsilon }_{5}$ 97.1

Table 2. Comparison of results obtained via different skeleton input data
View table
View in Article
Table 2. Comparison of results obtained via different skeleton input data
结构数据 CV精度/%
三流骨架、关节、动态 97.1
w/o骨架 96.0
两流 w/o关节 95.8
w/o动态 95.2
骨架 93.7
单流关节 93.5
动态 92.1

Table 3. The performance of models with different dilation factors
View table
View in Article
Table 3. The performance of models with different dilation factors
空洞率 CV精度/%
1 95.1
2 95.6
3 96.1
4 95.9
1, 2, 3, 4 97.1

Table 4. Comparison with state-of-the-art methods on the NTU-RGB+D dataset

View table

View in Article

Table 4. Comparison with state-of-the-art methods on the NTU-RGB+D dataset

方法	CS精度/%	CV精度/%
Lie-Group^[2]	50.1	52.8
TCN^[23]	74.3	83.1
ST-GCN^[9]	86.8	94.2
AS-GCN^[14]	86.8	94.2
2s-AGCN^[13]	88.5	95.1
SGN^[17]	89.0	94.5
AGC-LSTM^[7]	89.2	95.0
DGNN^[24]	89.9	96.1
Res-GCN^[11]	90.0	96.0
SGCN^[16]	90.1	96.2
MHCN	91.2	97.1

Table 5. Comparison with state-of-the-art methods on the Kinetics dataset
View table
View in Article
Table 5. Comparison with state-of-the-art methods on the Kinetics dataset
方法 TOP1精度/% TOP5精度/%
TCN^[23] 20.3 40.0
ST-GCN^[9] 30.7 52.8
AS-GCN^[14] 34.8 56.5
DGNN^[24] 36.9 59.6
2s-AGCN^[13] 36.1 58.7
Hyper-GCN^[18] 37.1 60.0
SGCN^[16] 37.1 60.1
MHCN 38.1 61.8

Tools

Get Citation

Copy Citation Text

Xiaofei QIN, Ying ZHAO, Yijie ZHANG, Ruijie DU, Hanwen QIAN, Meng CHEN, Wenqi ZHANG, Xuedian ZHANG. Multiscale hypergraph convolutional network for skeleton-based action recognition[J]. Optical Instruments, 2022, 44(4): 39

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites