Table 7 Comparative experimental results of different algorithms.
From: Improved 3D-ResNet sign language recognition algorithm with enhanced hand features
Core method/network | References | Dataset | General class | Accuracy (%) |
|---|---|---|---|---|
I3D+BLSTM | Adaloglou et al.30 | GSL isol | 310 | 89.74 |
B3D ResNet | Liao et al.31 | DEVISIGN_D | 500 | 89.80 |
AM-ResC3D+RCNN | Zhang et al.32 | DEVISIGN_D | 500 | 91.00 |
Method of Fakhfakh et al | Fakhfakh et al.33 | SIGNUM | 450 | 90.10 |
Method of Xiao et al | Xiao et al.34 | Kinect RGB-D | 500 | 85.24 ± 1.83 |
Ours | Â | CSL | 500 | 91.12 |
