Skip to main content
. 2018 May 31;41(7):1559–1572. doi: 10.1109/TPAMI.2018.2840695

Fig. 3.

Fig. 3.

The CNN structure of 2D/3D P-Net with CRF-Net(f). The parameters of convolution layers are (kernel size, output channels, dilation) in dark blue rectangles. Block 1 to block 6 are resolution-preserving. 2D/3D R-Net uses the same structure as 2D/3D P-Net except its input has three additional channels shown in Fig. 2 and the CRF-Net(f) is replaced by the CRF-Net(fu) (Section 3.3).