Detection Performance Evaluation of Boosted Random Ferns

Detection Performance Evaluation of Boosted Random Ferns. M. Villamizar, F. Moreno-Noguer, J. Andrade-Cetto, A. Sanfeliu. Iberian Conference on Pattern Recognition and Image Analysis (IBPRIA), 2011. [Results on the INRIA horses dataset] Description: We show that adding an iterative bootstrapping phase during the learning of the object classifier, it increases its detection rates given that additional positive and negative samples are collected (bootstrapped) for retraining the boosted classifier. After each bootstrapping iteration, the learning algorithm is concentrated on computing more discriminative and robust features (Random Ferns), since the bootstrapped samples extend the training data with more difficult images. Green rectangles indicate correct detections -true positives-, whereas red rectangles are false positives. Links: http://www.iri.upc.edu/people/mvillam... Contact: Michael Villamizar mvillami-at-iri.upc.edu Institut de Robòtica i Informática Industrial CSIC-UPC Barcelona - Spain