Research Article
[Retracted] Rethinking Separable Convolutional Encoders for End-to-End Semantic Image Segmentation
Table 7
Structure details of the discriminant network.
| Operation | Convolution kernel size | Input channel | Output channel | Step | Batch normalization |
| Conv 1 | 4∗4 | 4 | 144 | 1 | N | Conv 2 | 4∗4 | 144 | 144 | 2 | Y | Conv 3 | 4∗4 | 144 | 288 | 1 | Y | Conv 4 | 4∗4 | 288 | 288 | 2 | Y | Conv 5 | 4∗4 | 288 | 432 | 1 | Y | Conv 6 | 4∗4 | 432 | 432 | 2 | Y | Conv 7 | 4∗4 | 432 | 576 | 1 | Y | Conv 8 | 4∗4 | 576 | 576 | 2 | Y |
|
|