Figure 2.

The overall 2D convolution operation. The input data dimension is $W\times H\times C_1$, where $C_1$ is the channel number; the 2D convolution kernel size is $\mathrm{k}\times \mathrm{k}$ and it denotes the coverage of convolution kernel over spatial dimension in each convolution operation; the output data dimension of single kernel is $\left(W-k+1\right)\left(H-k+1\right)$ and the final output data generated by $\mathrm{p}$ kernels is a 3D tensor.