FAOD-Net: A Fast AOD-Net for Dehazing Single Image

Figure 1

(a) The process of standard convolution. (b) The process of depthwise convolution. (c) The process of pointwise convolution. H, W, and C are the height, width, and number of channels of the input feature map P, respectively. N is the number of filters, and k is the convolution kernel size.

(a)

(b)

(c)