International Journal of Computer Vision

, Volume 110, Issue 3, pp 259–274

People Watching: Human Actions as a Cue for Single View Geometry

  • David F. Fouhey
  • Vincent Delaitre
  • Abhinav Gupta
  • Alexei A. Efros
  • Ivan Laptev
  • Josef Sivic
Article

DOI: 10.1007/s11263-014-0710-z

Cite this article as:
Fouhey, D.F., Delaitre, V., Gupta, A. et al. Int J Comput Vis (2014) 110: 259. doi:10.1007/s11263-014-0710-z

Abstract

We present an approach which exploits the coupling between human actions and scene geometry to use human pose as a cue for single-view 3D scene understanding. Our method builds upon recent advances in still-image pose estimation to extract functional and geometric constraints on the scene. These constraints are then used to improve single-view 3D scene understanding approaches. The proposed method is validated on monocular time-lapse sequences from YouTube and still images of indoor scenes gathered from the Internet. We demonstrate that observing people performing different actions can significantly improve estimates of 3D scene geometry.

Keywords

Scene understandingAction recognition3D reconstruction

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • David F. Fouhey
    • 1
  • Vincent Delaitre
    • 2
  • Abhinav Gupta
    • 1
  • Alexei A. Efros
    • 1
    • 3
  • Ivan Laptev
    • 2
  • Josef Sivic
    • 2
  1. 1.Robotics InstituteCarnegie Mellon UniversityPittsburghUSA
  2. 2.WILLOW Project, Département d’Informatique de l’École Normale SupérieureENS/INRIA/CNRS UMR 8548ParisFrance
  3. 3.EECS Department at UC BerkeleyBerkeleyUSA