• DocumentCode
    2507868
  • Title

    Package Boosting for Readaption of Cascaded Classifiers

  • Author

    Szczot, Magdalena ; Forster, Julian ; Löhlein, Otto ; Palm, Günther

  • Author_Institution
    Dept. Environ. Perception (GR/PAP), Daimler AG, Ulm, Germany
  • fYear
    2010
  • fDate
    23-26 Aug. 2010
  • Firstpage
    552
  • Lastpage
    555
  • Abstract
    This contribution presents an efficient and useful way to readapt a cascaded classifier. We introduce Package Boosting which combines the advantages of Real Adaboost and Online Boosting for the realization of the strong learners in each cascade layer. We also examine the conditions which need to be fulfilled by a cascade in order to meet the requirements of an online algorithm and present the evaluation results of the system.
  • Keywords
    learning (artificial intelligence); pattern classification; cascaded classifiers; online boosting; package boosting; real Adaboost; Approximation algorithms; Boosting; Classification algorithms; Detectors; Equations; Estimation; Training; Classifier Readaption; Online Boosting;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Pattern Recognition (ICPR), 2010 20th International Conference on
  • Conference_Location
    Istanbul
  • ISSN
    1051-4651
  • Print_ISBN
    978-1-4244-7542-1
  • Type

    conf

  • DOI
    10.1109/ICPR.2010.140
  • Filename
    5597441