VAE/GAN

the tensorflow code of Autoencoding beyond pixels using a learned similarity metric

The paper should be the first one to combine the Variational Autoencoder(VAE) and Generative Adversarial Networks(GAN), by using the discrimiator of GAN as the perceptual loss instead of the pixel-wise loss in the original VAE. VAE/GAN also can be used for image reconstruction and visual attribution manipulation.

Prerequisites

• tensorflow >=1.0

dataset requirement

You can download the CelebA dataset and unzip CelebA into a directory. Noted that this directory don't contain the sub-directory.

Usage

Train:

$ python main.py --operation 0 --path your data path

Test:

$ python main.py --operation 1 --path your data path

##Experiments visual result

Sampling

Reconstruction(Different cropping size.)

##Issue If you find the bug and problem, Thanks for your issue to propose it.

Reference code

DCGAN

autoencoding_beyond_pixels