Skip to main content
. Author manuscript; available in PMC: 2020 May 31.
Published in final edited form as: Proc SPIE Int Soc Opt Eng. 2020 Mar 10;11313:1131313. doi: 10.1117/12.2549716

Figure 1.

Figure 1.

CNN architectures. (a) Model I: four-level U-Net with two input channels, and (b) Model II: four-level DIP U-Net with two contraction paths. The number of feature maps in each layer is mentioned on the top of that layer. For each level, the size of each feature map is mentioned at the left side of the level.