云鉴权>> FaceGen-GAN>> 返回
项目作者： alexpod1000

项目描述 ：
Conditional face generation experiments using GAN models on CelebA dataset.
高级语言： Python
项目地址: git://github.com/alexpod1000/FaceGen-GAN.git
创建时间： 2020-07-20T16:57:41Z
项目社区：https://github.com/alexpod1000/FaceGen-GAN
开源协议：MIT License
下载


FaceGen-GAN

Conditional face generation experiments using GAN models on CelebA dataset.

Architectures

  • Vanilla DCGAN: a normal DCGAN as described in DCGAN paper, has training stability issues.
  • Hinge DCGAN with custom layers: an improved DCGAN with spectral normalization, self-attention, minibatch std and pixelwise normalization, which allows stable training with better visual results than DCGAN.

EMA Training



In order to improve generated image quality, it is also possible to train a model using exponential moving average (EMA) update, as defined in The Unusual Effectiveness of Averaging in GAN Training paper.

The code is based on the update function found here, which allows updating a second generator model’s weights using EMA update using the following equation:

wt+1 = (1 - b) ut + (b) wt (assuming u are the weights of a generator trained via gradient methods)