Integrating Multiple Viewpoints for Articulated Scene Model Aquisition

  • Leon Ziegler
  • Agnes Swadzba
  • Sven Wachsmuth
Conference paper

DOI: 10.1007/978-3-642-39402-7_30

Part of the Lecture Notes in Computer Science book series (LNCS, volume 7963)
Cite this paper as:
Ziegler L., Swadzba A., Wachsmuth S. (2013) Integrating Multiple Viewpoints for Articulated Scene Model Aquisition. In: Chen M., Leibe B., Neumann B. (eds) Computer Vision Systems. ICVS 2013. Lecture Notes in Computer Science, vol 7963. Springer, Berlin, Heidelberg

Abstract

In this paper we present a method to generate an Articulated Scene Model for a system’s current view, which allows to integrate multiple egocentric models previously gathered from different viewpoints. The approach is designed to build up separate representations for the static, movable and dynamic parts of an observed scene. In order to make already gathered information available for subsequent viewpoints of the same location, a merging algorithm is needed that considers view-dependent aspects like occlusion and limitations of the view frustum. We show in our experiments that the proposed algorithm correctly merges multiple scene models and can be applied profitably in an integrated vision system for detecting movable objects on a mobile robot.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Leon Ziegler
    • 1
  • Agnes Swadzba
    • 1
  • Sven Wachsmuth
    • 1
    • 2
  1. 1.Applied Informatics, Faculty of TechnologyBielefeld UniversityGermany
  2. 2.Central Lab Facilities, CITECBielefeldGermany

Personalised recommendations