A Semi-supervised Deep Generative Model for Human Body Analysis

Rodrigo de Bem, Arnab Ghosh, Thalaiyasingam Ajanthan, Ondrej Miksik, N. Siddharth, Philip Torr

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Deep generative modelling for human body analysis is an emerging problem with many interesting applications. However, the latent space learned by such models is typically not interpretable, resulting in less flexible models. In this work, we adopt a structured semi-supervised approach and present a deep generative model for human body analysis where the body pose and the visual appearance are disentangled in the latent space. Such a disentanglement allows independent manipulation of pose and appearance, and hence enables applications such as pose-transfer without being explicitly trained for such a task. In addition, our setting allows for semi-supervised pose estimation, relaxing the need for labelled data. We demonstrate the capabilities of our generative model on the Human3.6M and on the DeepFashion datasets.
Original languageEnglish
Title of host publicationComputer Vision -- ECCV 2018 Workshops
EditorsLaura Leal-Taixé, Stefan Roth
Place of PublicationCham
PublisherSpringer International Publishing
Number of pages18
ISBN (Electronic)978-3-030-11012-3
ISBN (Print)978-3-030-11011-6
Publication statusPublished - 29 Jan 2019
Event9th International Workshop on Human Behavior Understanding: Generating Visual Data of Human Behavior - Munich, Germany
Duration: 9 Sep 20189 Sep 2018

Publication series

NameLecture Notes in Computer Science
PublisherSpringer, Cham
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Workshop9th International Workshop on Human Behavior Understanding
Abbreviated titleHBUGEN 2018
Internet address


  • Deep generative models
  • Variational autoencoders
  • Semi-supervised learning
  • Human pose estimation
  • Analysis-by-synthesis

Cite this