IS-GAN:Learning Disentangled Representation for Robust Person Re-identification

Chanho Eom, Wonkyung Lee, Geon Lee, Bumsub Ham

Research output: Contribution to journalArticlepeer-review


We address the problem of person re-identification (reID), that is, retrieving person images from a large dataset, given a query image of the person of interest. A key challenge is to learn person representations robust to intra-class variations, as different persons could have the same attribute, and persons appearances look different, e.g., with viewpoint changes. Recent reID methods focus on learning person features discriminative only for a particular factor of variations, which also requires corresponding supervisory signals. To tackle this problem, we propose to factorize person images into identity-related and -unrelated features. Identity-related features contain information useful for specifying a particular person, while identity-unrelated ones hold other factors. To this end, we propose a new generative adversarial network, dubbed IS-GAN. It disentangles identity-related and -unrelated features through an identity-shuffling technique that exploits identification labels alone without any auxiliary supervisory signals. We restrict the distribution of identity-unrelated features, or encourage identity-related and -unrelated features to be uncorrelated, facilitating the disentanglement process. Experimental results validate the effectiveness of IS-GAN, showing state-of-the-art performance on standard reID benchmarks. We further demonstrate the advantages of disentangling person representations on a long-term reID task, setting a new state of the art on a Celeb-reID dataset.

Original languageEnglish
JournalIEEE transactions on pattern analysis and machine intelligence
Publication statusAccepted/In press - 2021

Bibliographical note

Publisher Copyright:

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics


Dive into the research topics of 'IS-GAN:Learning Disentangled Representation for Robust Person Re-identification'. Together they form a unique fingerprint.

Cite this