TNT logo Zur zentralen Website der Leibniz Universität Hannover Bannerlogo des Instituts für Informationsveratbeitung

Multiple People Tracking

TNT members involved in this project:
Dipl.-Math. Roberto Henschel
Prof. Dr.-Ing. Bodo Rosenhahn
Show all

Multiple people tracking is a challenging task of the computer vision domain, with applications in surveillance, action recognition and for example human-computer interaction systems.

Whereas single object tracking can be considered as almost solved, multiple people tracking still remains a difficult problem with ongoing research, due to the complex interaction between persons, that needs to be modeled in order to obtain good results.

We are interested in finding the global optimal solution to the data association problem. In order to do so, we propose to solve the tracking problem by using knowledge from graph theory:

  • We model the association problem as a hierarchical tracking problem in a DAG, which reduces wrong decisions in each step compared to other hierarchical approaches and we solve the association problem by efficiently computing a minimum cost arborescence.
  • We model the association problem in a network flow graph, include social and grouping behavior between objects and solve it by applying the simplex algorithm.

Show recent publications only
  • Book Chapters
    • Laura Leal-Taixé, Bodo Rosenhahn
      Pedestrian interaction in tracking: the social force model and global optimization methods
      Modeling, Simulation and Visual Analysis of Crowds: A multidisciplinary perspective, Springer, September 2012, edited by Saad Ali, Ko Nishino, Dinesh Manocha and Mubarak Shah
    • Laura Leal-Taixé, Gerard Pons-Moll, Bodo Rosenhahn
      Exploiting pedestrian interaction via global optimization and social behaviors
      Theoretic Foundations of Computer Vision: Outdoor and Large-Scale Real-World Scene Analysis, Springer, April 2012, edited by F. Dellaert, J.-M. Frahm, M. Pollefeys, L. Leal-Taixé, B. Rosenhahn
  • Conference Contributions
    • Roberto Henschel, Laura Leal-Taixé, Daniel Cremers, Bodo Rosenhahn
      Fusion of Head and Full-Body Detectors for Multi-Object Tracking
      Computer Vision and Pattern Recognition Workshops (CVPRW), accepted as spotlight presentation, June 2018
    • Holger Meuel, Luis Angerstein, Roberto Henschel, Bodo Rosenhahn, Jörn Ostermann
      Moving Object Tracking for Aerial Video Coding using Linear Motion Prediction and Block Matching
      Proceedings of the 32nd Picture Coding Symposium (PCS), pp. 1-5, Nurmberg, Germany, December 2016
    • Roberto Henschel, Laura Leal-Taixé, Rosenhahn Bodo
      Solving Multiple People Tracking In A Minimum Cost Arborescence
      IEEE Winter Conference on Applications of Computer Vision Workshops (WACVW), accepted as oral presentation, 1st Workshop on Benchmarking Multi-target Tracking (BMTT), Waikoloa Beach, Hawaii, USA, January 2015
    • Roberto Henschel, Laura Leal-Taixé, Bodo Rosenhahn
      Efficient Multiple People Tracking Using Minimum Cost Arborescences
      German Conference on Pattern Recognition (GCPR), accepted as oral presentation, Münster, Germany, September 2014
    • Laura Leal-Taixé, Michele Fenzi, Alina Kuznetsova, Bodo Rosenhahn, Silvio Savarese
      Learning an Image-based Motion Context for Multiple People Tracking
      IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, Ohio, USA, June 2014
    • Laura Leal-Taixé, Gerard Pons-Moll, Bodo Rosenhahn
      Branch-and-price global optimization for multi-view multi-object tracking
      IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Providence, Rhode Island, USA., June 2012
    • Laura Leal-Taixé, Gerard Pons-Moll, Bodo Rosenhahn
      Everybody needs somebody: modeling social and grouping behavior on a linear programming multiple people tracker
      IEEE International Conference on Computer Vision Workshops (ICCVW). 1st Workshop on Modeling, Simulation and Visual Analysis of Large Crowds, November 2011
  • Technical Report
    • Roberto Henschel, Laura Leal-Taixé, Bodo Rosenhahn, Konrad Schindler
      Tracking with multi-level features
      arXiv, July 2016