Fig. 4: Network architecture of the proposed U-Net.

Include an encoder, a bottleneck layer, and a decoder. Conv denotes a convolutional layer, while BN indicates batch normalization. The sizes of feature maps of each layer are shown around each box.
Include an encoder, a bottleneck layer, and a decoder. Conv denotes a convolutional layer, while BN indicates batch normalization. The sizes of feature maps of each layer are shown around each box.