Dr. Barbara Krausz


I am a researcher in NetMedia's computer vision group at Fraunhofer IAIS. I received my bachelor's and master's degree from the Bonn-Rhein-Sieg University of Applied Sciences in Sankt Augustin and my PhD in 2012 from the University of Bonn. In 2008, I started to work at Fraunhofer IAIS.

Research Interests

  • event detection and action recognition in videos
  • crowd dynamics
  • pedestrian simulation
  • object detection and tracking
  • multimedia analysis and retrieval
  • image aesthetics


  • B. Krausz, C. Bauckhage
    Integrating lateral swaying of pedestrians into simulations, Int. Conf. on Pedestrian and Evacuation Dynamics PED 2012
  • D. Stein, B. Krausz, J. Löffler, R. Bardeli, J. Schwenninger, B. Usabaev, R. Marterer.
    Enriching an Intelligent Resource Management System with Automatic Event Recognition
    ISCRAM 2012
  • B. Krausz, C. Bauckhage
    Loveparade 2010: Automatic Video Analysis of a Crowd Disaster
    Computer Vision and Image Understanding, 116(3), 307 - 319, 2012
  • B. Krausz, C. Bauckhage
    Analyzing Pedestrian Behavior in Crowds for Automatic Detection of Congestions
    ICCV-Workshop on Modeling, Simulation and Visual Analysis of Large Crowds, 2011
  • B. Krausz, C. Bauckhage
    Automatic Detection of Dangerous Motion Behavior in Human Crowds
    AVSS 2011
  • B. Krausz, C. Bauckhage
    Action Recognition in Videos using Nonnegative Tensor Factorization
    ICPR 2010
  • B. Krausz, R. Herpers
    MetroSurv: detecting events in subway stations,
    Multimedia Tools and Applications, 50 (1), S. 123-147, 2010
  • B. Krausz, R. Herpers
    Event detection for video surveillance using an expert system,
    AREA 2008
  • B. Krausz, A. Brièll, C. Eckes, J. Löffler
    Capturing and Processing of 360° Panoramic Images for Emergency Site Exploration,
    Mobile Response 2008

In Media

  • heute - in Deutschland, hallo deutschland, ZDF, 08.09.2011 [Mediathek]
  • WDR aktuell, WDR, 06.09.2011
  • Software verhindert künftig Massenpaniken, computerworld.ch, 12.08.2011 [www]
  • Love Parade: Software sucht Hotspots in Menschenmassen, pressetext, 10.08.2011 [www]
  • Smart software spots swaying risk of a crushing crowd, New Scientist, 09.08.2011 [www]
  • Festival crush disasters could be prevented with software that detects dangerous crowd build-up, Daily Mail Online, 04.08.2011 [www]
  • Gefahr bei Großveranstaltungen: Software erkennt Vorboten einer Massenpanik, SpiegelONLINE, 26.11.2010 [www]
  • Loveparade-Videos analysiert, Kölnische Rundschau, 06.10.2010
  • Software gegen Massenpanik, Kölner Stadt-Anzeiger, 01.10.2010 [www]

Latest blog entries

Foto des Autors

CVIU Article

New article about the Loveparade stampede in Computer Vision and Image Understanding.


ZDF and WDR report on our video analysis system in "WDR aktuell", "heute - in Deutschland" and "hallo deutschland"


Our paper "Analyzing Pedestrian Behavior in Crowds for Automatic Detection of Congestions" has been accepted for the ICCV Workshop on Modeling, Simulation and Visual Analysis of Large Crowds.


NewScientist reports on our work to detect dangerous motion behavior in crowds