Disentangling by Factorising

Code

• github.com/clementchadebec/benchmark_VAE
  pytorch★ 1,983
• github.com/facebookresearch/disentangling-correlated-factors
  pytorch★ 78
• github.com/mmrl/disent-and-gen
  pytorch★ 21
• github.com/AliLotfi92/Disentangling_by_Factorising
  pytorch★ 0
• github.com/1Konny/FactorVAE
  pytorch★ 0
• github.com/danielbraithwt/Readings
  none★ 0
• github.com/nicolasigor/FactorVAE
  tf★ 0
• github.com/gene-chou/conditional-factor-vae
  pytorch★ 0
• github.com/wangdedi1997/Disentanglement-Beta-FactorVAE
  tf★ 0
• github.com/Guiliang/FactorVAE
  pytorch★ 0

Abstract

We define and address the problem of unsupervised learning of disentangled representations on data generated from independent factors of variation. We propose FactorVAE, a method that disentangles by encouraging the distribution of representations to be factorial and hence independent across the dimensions. We show that it improves upon -VAE by providing a better trade-off between disentanglement and reconstruction quality. Moreover, we highlight the problems of a commonly used disentanglement metric and introduce a new metric that does not suffer from them.

Tasks

Reproductions