Table 1 Network architecture of our proposed method
From: Facial attribute-controlled sketch-to-image translation with generative adversarial networks
Generator | Discriminator | Attribute classifier | |
|---|---|---|---|
Conv(64, 5, 1) | Conv(64, 3, 2) | ||
Conv(128, 3, 2) | Conv(128, 3, 2) | ||
Conv(256, 3, 2) | Conv(256, 3, 2) | ||
Residual blocks | Conv(256, 3, 1) | Conv(512, 3, 2) | |
Conv(256, 3, 1) | Conv(1024, 3, 2) | ||
Conv(256, 3, 1) | FC(1024) | FC(1024) | |
Conv(256, 3, 1) | FC(1) | FC(7) | |
Conv(256, 3, 1) | Â | Â | |
Conv(256, 3, 1) | Â | Â | |
DeConv(128, 3, 2) | Â | Â | |
DeConv(64, 3, 2) | Â | Â | |
DeConv(3, 5, 1) | Â | Â | |