Disentangled Representation Learning Using ($\beta$-)VAE and GAN

TL;DR

This paper explores how combining ($\beta$-)VAE with GANs can produce disentangled representations of image features like shape, size, and position, enhancing interpretability and image quality.

Contribution

It introduces a method that integrates VAE and GAN to achieve disentangled feature representations while improving image reconstruction quality.

Findings

Disentangled features were successfully isolated in the hidden space.

GAN improved the visual quality of reconstructed images.

Disruption of each dimension revealed meaningful feature encodings.

Abstract

Given a dataset of images containing different objects with different features such as shape, size, rotation, and x-y position; and a Variational Autoencoder (VAE); creating a disentangled encoding of these features in the hidden space vector of the VAE was the task of interest in this paper. The dSprite dataset provided the desired features for the required experiments in this research. After training the VAE combined with a Generative Adversarial Network (GAN), each dimension of the hidden vector was disrupted to explore the disentanglement in each dimension. Note that the GAN was used to improve the quality of output image reconstruction.