Affordable Access

Access to the full text

An Adversarial Neuro-Tensorial Approach for Learning Disentangled Representations

Authors
  • Wang, Mengjiao1
  • Shu, Zhixin2
  • Cheng, Shiyang1
  • Panagakis, Yannis1, 3
  • Samaras, Dimitris2
  • Zafeiriou, Stefanos1
  • 1 Imperial College London, London, UK , London (United Kingdom)
  • 2 Stony Brook University, Stony Brook, USA , Stony Brook (United States)
  • 3 Middlesex University, London, UK , London (United Kingdom)
Type
Published Article
Journal
International Journal of Computer Vision
Publisher
Springer-Verlag
Publication Date
Feb 16, 2019
Volume
127
Issue
6-7
Pages
743–762
Identifiers
DOI: 10.1007/s11263-019-01163-7
Source
Springer Nature
Keywords
License
Green

Abstract

Several factors contribute to the appearance of an object in a visual scene, including pose, illumination, and deformation, among others. Each factor accounts for a source of variability in the data, while the multiplicative interactions of these factors emulate the entangled variability, giving rise to the rich structure of visual object appearance. Disentangling such unobserved factors from visual data is a challenging task, especially when the data have been captured in uncontrolled recording conditions (also referred to as “in-the-wild”) and label information is not available. In this paper, we propose a pseudo-supervised deep learning method for disentangling multiple latent factors of variation in face images captured in-the-wild. To this end, we propose a deep latent variable model, where the multiplicative interactions of multiple latent factors of variation are explicitly modelled by means of multilinear (tensor) structure. We demonstrate that the proposed approach indeed learns disentangled representations of facial expressions and pose, which can be used in various applications, including face editing, as well as 3D face reconstruction and classification of facial expression, identity and pose.

Report this publication

Statistics

Seen <100 times