zhangqianhui / Vae Gan Tensorflow
Tensorflow code of "autoencoding beyond pixels using a learned similarity metric"
Programming Languages
python 139335 projects - #7 most used programming language
Projects that are alternatives of or similar to Vae Gan Tensorflow
ExermoteUsing Machine Learning to predict the type of exercise from movement data
Stars: ✭ 108 (-6.9%)
Mutual labels: gan
DeepnudecliDeepNude Command Line Version With Watermark Removed
Stars: ✭ 112 (-3.45%)
Mutual labels: gan
FaganA variant of the Self Attention GAN named: FAGAN (Full Attention GAN)
Stars: ✭ 105 (-9.48%)
Mutual labels: gan
Convolutional Handwriting GanScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation (CVPR20)
Stars: ✭ 107 (-7.76%)
Mutual labels: gan
Nmt gangenerative adversarial nets for neural machine translation
Stars: ✭ 110 (-5.17%)
Mutual labels: gan
Zerospeech Tts Without TA Pytorch implementation for the ZeroSpeech 2019 challenge.
Stars: ✭ 100 (-13.79%)
Mutual labels: gan
ImpersonatorPyTorch implementation of our ICCV 2019 paper: Liquid Warping GAN: A Unified Framework for Human Motion Imitation, Appearance Transfer and Novel View Synthesis
Stars: ✭ 1,605 (+1283.62%)
Mutual labels: gan
PsganPeriodic Spatial Generative Adversarial Networks
Stars: ✭ 108 (-6.9%)
Mutual labels: gan
PokeganCreats new kinds of pokemons using WGAN!
Stars: ✭ 112 (-3.45%)
Mutual labels: gan
3dgan Pytorch3DGAN-Pytorch
Stars: ✭ 105 (-9.48%)
Mutual labels: gan
Tf Exercise GanTensorflow implementation of different GANs and their comparisions
Stars: ✭ 110 (-5.17%)
Mutual labels: gan
TbganProject Page of 'Synthesizing Coupled 3D Face Modalities by Trunk-Branch Generative Adversarial Networks'
Stars: ✭ 105 (-9.48%)
Mutual labels: gan
Sketch To Art🖼 Create artwork from your casual sketch with GAN and style transfer
Stars: ✭ 115 (-0.86%)
Mutual labels: gan
What I Have ReadPaper Lists, Notes and Slides, Focus on NLP. For summarization, please refer to https://github.com/xcfcode/Summarization-Papers
Stars: ✭ 110 (-5.17%)
Mutual labels: gan
Msg Gan V1MSG-GAN: Multi-Scale Gradients GAN (Architecture inspired from ProGAN but doesn't use layer-wise growing)
Stars: ✭ 116 (+0%)
Mutual labels: gan
Hccg CycleganHandwritten Chinese Characters Generation
Stars: ✭ 115 (-0.86%)
Mutual labels: gan
GpndGenerative Probabilistic Novelty Detection with Adversarial Autoencoders
Stars: ✭ 112 (-3.45%)
Mutual labels: gan
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.
About training instability
I also found the training is very instability. So, I update the code to stablize the adversarial progress of VAE/GAN. The details is in the below.
Pretrained models.
The checkpoints files can be downloads from Google Drive. Please unzip the files inside the project directory. Later, I will update the new models after more training iterations.
Prerequisites
dataset requirement
You can download the Align and Cropped CelebA dataset
and unzip CelebA into a directory. Noted that this directory don't contain the sub-directory.
Usage
Train:
$ python main.py --op 0 --path your data path
Test:
$ python main.py --op 1 --path your data path
Experiments visual result
Input:
Reconstruction
Issue
If you find the bug and problem, Thanks for your issue to propose it.
Reference code
DCGAN
autoencoding_beyond_pixels
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at
[email protected].