Learning Disentangled Representations with Reference-Based Variational Autoencoders

Abstract : Learning disentangled representations from visual data, where different high-level generative factors are independently encoded, is of importance for many computer vision tasks. Solving this problem, however, typically requires to explicitly label all the factors of interest in training images. To alleviate the annotation cost, we introduce a learning setting which we refer to as reference-based disentangling. Given a pool of unlabelled images, the goal is to learn a representation where a set of target factors are disentangled from others. The only supervision comes from an auxiliary reference set containing images where the factors of interest are constant. In order to address this problem, we propose reference-based variational autoencoders, a novel deep generative model designed to exploit the weak-supervision provided by the reference set. By addressing tasks such as feature learning, conditional image generation or attribute transfer, we validate the ability of the proposed model to learn disentangled representations from this minimal form of supervision.
Complete list of metadatas

Cited literature [18 references]  Display  Hide  Download

https://hal.inria.fr/hal-01896007
Contributor : Adrià Ruiz <>
Submitted on : Wednesday, January 23, 2019 - 12:47:11 PM
Last modification on : Thursday, September 12, 2019 - 2:48:31 PM

File

RbVAES_HAL.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01896007, version 2

Collections

Citation

Adrià Ruiz, Oriol Martinez, Xavier Binefa, Jakob Verbeek. Learning Disentangled Representations with Reference-Based Variational Autoencoders. ICLR 2019 - 7th International Conference on Learning Representations, May 2019, New Orleans, United States. pp.1-17. ⟨hal-01896007v2⟩

Share

Metrics

Record views

174

Files downloads

906