Skip to main content

Table 1 Structure of discriminator

From: Dynamic scene deblurring and image de-raining based on generative adversarial networks and transfer learning for Internet of vehicle

#

Layer

Size of feature map

Stride

Parameters

Receptive field

1

Input

\(256 \times 256 \times 3\)

–

–

–

2

Conv

\(128 \times 128 \times 64\)

2

3136

142

3

Conv

\(64 \times 64 \times 64\)

2

65,600

70

4

Conv

\(32 \times 32 \times 128\)

2

131,200

34

5

Conv

\(16 \times 16 \times 256\)

2

524,544

16

6

Conv

\(16 \times 16 \times 512\)

1

2,097,644

7

7

Conv

\(16 \times 16 \times 1\)

1

8193

4

8

Dense

–

–

263,168

–

9

Dense

–

–

1025

–