Table 7 Comparison of our proposed DSwarm-Net model with existing methods on the HDM05 dataset using 10 random split-mean protocol.

From: A union of deep learning and swarm-based optimization for 3D human action recognition

Method

Year

Classification accuracy (%)

HCN72

2018

86.51

PB-GCN26

2018

88.2

Deep STGC27

2019

85.29

2S-AGCN29

2019

88.5

PGCN-TCA30

2020

86.71

SGCN73

2021

85.45

Di-StddNet74

2021

82.32

DSwarm-Net

2021

90.67