Research Article
An Indirect Multimodal Image Registration and Completion Method Guided by Image Synthesis
Figure 1
The overall framework of our proposed ACE-GAN. It consists of four subnetworks, encoder (E), generator (G), discriminator (D), and classifier (C). The detailed structure of the encoder and decoder residual block is shown in (a) and (b), respectively. Specifically, s1k1 represents that the convolution stride is 1 and the kernel size is .