Various calculations for Convolutions

https://www.slideshare.net/ssuser06e0c5/convolution-77257148

1. Common convolution

input feature map : F x F x N(number of channels)

output feature map : F x F x M(number of channels)

=> 

requirement: 

numbers of kernel parameters : K x K x N x M(number of kernels)

total calculations : F x F x N x K x K x M

2. Pointwise convolution

input feature map : F x F x N(number of channels)

output feature map : F x F x M(number of channels)

=>

requirement: 

number of kernel parameters : 1 x 1 x N x M

total calculations: F x F x 1 x 1 x N x M

3. Depthwise convolution

input feature map : F x F x N(number of channels)

output feature map : F x F x N(number of channels)

=>

requirement: 

number of kernel parameters : K x K x N

total calculations: F x F x N x K x K