Person Detection with a Computation Time Weighted AdaBoost
Résumé
In this paper, a boosted cascade person detection framework with heterogeneous pool of features is presented. The boosted cascade construction and feature selection is carried out using a modified AdaBoost that takes computation time of features into consideration. The final detector achieves a low Miss Rate of 0.06 at 10−3 False Positive Per Window on the INRIA public dataset while achieving an average speed up of 1.8x on the classical variant.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...