Skip to main content
. Author manuscript; available in PMC: 2020 Oct 19.
Published in final edited form as: Med Image Anal. 2018 May 22;48:95–106. doi: 10.1016/j.media.2018.05.008

Fig. 3.

Fig. 3.

U-Net module The input image is of size 256 × 256 × N, where N is the number of channels (1 for all networks). Blue, green, and orange boxes correspond to multichannel feature maps. Green indicates a downsampled feature map. Orange indicates the result of a copy, concatenated with an upsampled feature map.