Disentangled human body embedding based on deep hierarchical neural network

Boyi Jiang, Juyong Zhang, Jianfei Cai, Jianmin Zheng

Research output: Contribution to journalArticleResearchpeer-review

11 Citations (Scopus)


Human bodies exhibit various shapes for different identities or poses, but the body shape has certain similarities in structure and thus can be embedded in a low-dimensional space. This paper presents an autoencoder-like network architecture to learn disentangled shape and pose embedding specifically for the 3D human body. This is inspired by recent progress of deformation-based latent representation learning. To improve the reconstruction accuracy, we propose a hierarchical reconstruction pipeline for the disentangling process and construct a large dataset of human body models with consistent connectivity for the learning of the neural network. Our learned embedding can not only achieve superior reconstruction accuracy but also provide great flexibility in 3D human body generation via interpolation, bilinear interpolation, and latent space sampling. The results from extensive experiments demonstrate the powerfulness of our learned 3D human body embedding in various applications.

Original languageEnglish
Pages (from-to)2560-2575
Number of pages16
JournalIEEE Transactions on Visualization and Computer Graphics
Issue number8
Publication statusPublished - 1 Aug 2020
Externally publishedYes


  • 3D Body Shape
  • 3D Human Articulated Body Model
  • Biological system modeling
  • Computational modeling
  • Deformation Representation
  • Face
  • Hierarchical Structure
  • Shape
  • Solid modeling
  • Strain
  • Three-dimensional displays
  • Variational Autoencoder

Cite this