Human-Computer Interaction with Hand Gesture Recognition Using ResNet and MobileNet

Table 3

Summary of the modified MobileNetV2 model.

Transfer learning	Pretrained on Google ImageNet dataset.
Number of layers	53 layers0 of MobileNetV2 architecture +3 sequential layers (2 fully connected layers + 1 softmax layer).Therefore, the total number of layers is 56 layers
Number of neurons in fully connected layers	1,024 neurons in the 1st fully connected layer, 512 neurons in the 2nd fully connected layer, and 32 neurons in the last layer (softmax layer)
Activation function	ReLu activation function
Optimizer	Adam optimizer with 0.0005 learning rate
Dropout	Dropout value used for the fully connected layers is 0.5
Batch size	32
Loss function	Categorical loss entropy
Number of epochs	15 epoch
Total no. of parameters	Total params: 8,043,104- trainable params: 8,008,992- nontrainable params: 34,112