Research Letters in the Information and Mathematical Sciences

Permanent URI for this collectionhttps://mro.massey.ac.nz/handle/10179/4332

Research Letters welcomes papers from staff and graduate students at Massey University in the areas of: Computer Science, Information Science, Mathematics, Statistics and the Physical and Engineering Sciences. Research letters is a preprint series that accepts articles of completed research work, technical reports, or preliminary results from ongoing research. After editing, articles are published online and can be referenced, or handed out at conferences. Copyright remains with the authors and the articles can be used as preprints to academic journal publications or handed out at conferences. Editors Dr Elena Calude Dr Napoleon Reyes The guidelines for writing a manuscript can be accessed here.

Browse

Filters

20101

Settings

Start date: 2010Subject: search.filters.subject.AdaBoost

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    A novel bootstrapping method for positive datasets in cascades of boosted ensembles
    (Massey University, 2010) Susnjak, T.; Barczak, A.L.C.; Hawick, K.A.
    We present a novel method for efficiently training a face detector using large positive datasets in a cascade of boosted ensembles. We extend the successful Viola-Jones [1] framework which achieved low false acceptance rates through bootstrapping negative samples with the capability to also bootstrap large positive datasets thereby capturing more in-class variation of the target object. We achieve this form of bootstrapping by way of an additional embedded cascade within each layer and term the new structure as the Bootstrapped Dual-Cascaded (BDC) framework. We demonstrate its ability to easily and efficiently train a classifier on large and complex face datasets which exhibit acute in-class variation.