相关资料
论文:
https://arxiv.org/abs/1812.04948

code :
https://github.com/NVlabs/stylegan
https://github.com/weihaox/awesome-gan-inversion
https://paperswithcode.com/paper/a-style-based-generator-architecture-for
https://github.com/huangzh13/StyleGAN.pytorch
https://github.com/rosinality/style-based-gan-pytorch

stylegan

The incremental list of changes to the generator are:

• Baseline Progressive GAN. 渐进式GAN
• Addition of tuning and bilinear upsampling. 添加微调共和双线性采样
• Addition of mapping network and AdaIN (styles). 添加映射网络和AdaIN (styles)
• Removal of latent vector input to generator.去除潜在向量输入到生成器。
• Addition of noise to each block.给每个模块添加噪音
• Addition Mixing regularization. 加法混合正则化。

  archive

  

参考:

https://www.geeksforgeeks.org/stylegan-style-generative-adversarial-networks/
https://machinelearningmastery.com/introduction-to-style-generative-adversarial-network-stylegan/

https://machinelearningmastery.com/introduction-to-style-generative-adversarial-network-stylegan/