BlendFace
BlendFace copied to clipboard
[ICCV 2023] BlendFace: Re-designing Identity Encoders for Face-Swapping https://arxiv.org/abs/2307.10854
self.Decoder_inchannel = [1024, 2048, 1024, 512, 256, 128] self.Decoder_outchannel = [1024, 512, 256, 128, 64, 32] self.Decoder = nn.ModuleDict({f'layer_{i}' : nn.Sequential( nn.ConvTranspose2d(self.Decoder_inchannel[i], self.Decoder_outchannel[i], kernel_size=4, stride=2, padding=1), nn.BatchNorm2d(self.Decoder_outchannel[i]), nn.LeakyReLU(0.1) )for i...
The swap result of this model is not good. The similarity of the source face with the swapped face is very poor. **Note that-> I have used proper alignment methods...
I want to know the code for computing 'Identity Distance' and 'Attribute Distance.' Thank you.
I notice that the result depends on how well the face is aligned, there is some script to align faces, it would be nice to be able to swap faces...
Could you tell me the total memory of mask.tar.gz after decompressing the file?