https://github.com/yzwxx/vae-celebA

Variational auto-encoder trained on celebA . All rights reserved.
https://github.com/yzwxx/vae-celebA

celeba-dataset tensorflow tensorlayer variational-autoencoder vggl16

Last synced: 6 days ago
JSON representation

Variational auto-encoder trained on celebA . All rights reserved.

• Host: GitHub
• URL: https://github.com/yzwxx/vae-celebA
• Owner: yzwxx
• Created: 2017-08-14T12:39:43.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2017-12-14T14:34:14.000Z (almost 7 years ago)
• Last Synced: 2024-08-02T20:44:33.318Z (3 months ago)
• Topics: celeba-dataset, tensorflow, tensorlayer, variational-autoencoder, vggl16
• Language: Python
• Size: 2.71 MB
• Stars: 181
• Watchers: 11
• Forks: 44
• Open Issues: 5
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-tensorlayer - Variational Autoencoder

README

        
# vae-celebA

Hereby we present plain VAE and modified VAE model, both of which are trained on celebA dataset to synthesize facial images.

## Result:

### plain VAE



      





      

  




      

  


### DFC-VAE

input image:  



      

  

reconstruction:  



      

  

randomly generation:  



      

  


To run the code, you are required to install Tensorflow and Tensorlayer on your machine. **[how to install Tensorlayer](https://github.com/zsdonghao/tensorlayer)**  

## SOME NOTES

This is the code for the paper **[Deep Feature Consistent Variational Autoencoder](https://houxianxu.github.io/assets/project/dfcvae)**  

In loss function we used a vgg loss.Check this **[how to load and use a pretrained VGG-16?](https://github.com/zsdonghao/tensorlayer/blob/master/example/tutorial_vgg16.py)** if you have trouble reading vgg_loss.py.  

## How to Run

Firstly, download the [celebA dataset](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html) and [VGG-16 weights](http://www.cs.toronto.edu/%7Efrossard/post/vgg16/).

After installing all the third-party packages required, we can train the models by:  

```python

python train_vae.py # for plain VAE

python train_dfc_vae.py # for DFC-VAE

```