shaohua0116 / VAE-Tensorflow

A Tensorflow implementation of a Variational Autoencoder for the deep learning course at the University of Southern California (USC).

Variational Autoencoder in Tensorflow

This is an Tensorflow implementation of a variational autoencoder for the deep learning course at USC (CSCI-599 Deep Learning and its Applications) taught by Professor Joseph Lim. The slides of this lecture are available here. This demo code is written by Shao-Hua Sun.

Results

Reconstruction

Generation

Transformation

Latent space

Related works

VAE

• The original VAE paper: Auto-Encoding Variational Bayes
• Variational Autoencoder: Intuition and Implementation
• Generating Large Images from Latent Vectors
• This demo code is partially based on the code from this post Variational Autoencoder in TensorFlow

Generative models

• My implementation of Semi-supervised learning GAN
• My implementation of Deep Convolutional GAN
• My implementation of Generative Latent Optimization

Author

Shao-Hua Sun / @shaohua0116 @ Joseph Lim's research lab @ USC