This paper demonstrates the value of improving the discriminating strength of weak classifiers in the context of boosting by using response binning. The reasoning is centered around, but not limited to, the well known Haar-features used by Viola & Jones in their face detection/pedestrian detection systems. It is shown that using a weak classifier based on a single threshold is sub-optimal and in the case of the Haar-feature inadequate. A more general method for features with multi-modal responses is derived that is easily used in boosting mechanisms that accepts a confidence measure, such as the RealBoost algorithm. The method is evaluated by boosting a single stage classifier and compare the performance to previous approaches.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Response Binning: Improved Weak Classifiers for Boosting


    Contributors:


    Publication date :

    2006-01-01


    Size :

    7313006 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Response Binning: Improved Weak Classifiers for Boosting

    Rasolzadeh, B. / Petersson, L. / Pettersson, N. | British Library Conference Proceedings | 2006


    Application of a fuzzy integral for weak classifiers boosting

    Samorodov, A. V. | British Library Online Contents | 2011


    Binning and non-binning combination

    YERUHAMI ODED / STEINBERG AMIT / OSIROFF NIR et al. | European Patent Office | 2024

    Free access