Table 3.

Comparison in terms of the number of parameters and the classification accuracy of the proposed model with other CNN models on UTD-MHAD dataset.

Model	Compact distance encoded features		Compact angle encoded features
	No. of parameter	Accuracy (%)	No. of parameter	Accuracy (%)
VGG19	57.9 M	83.45	105.7 M	88.45
VGG16	52.6 M	85.78	97.8 M	91.21
ResNet101	42.7 M	88.94	42.7 M	94.32
Inception v3	24.3 M	93.28	24.3 M	95.77
DenseNet201	18.4 M	92.64	18.4 M	95.24
DSwarm-Net	21.8 M	97.56	21.8 M	98.33