Framework for Research on Detection Classifiers

Framework for Research on Detection Classifiers

Paper in proceedings outside WoS and Scopus

Detection of patterns in images withclassifiers is currently one of the most important research topics in computervision. Many practical applications such as face detection exist and recentwork even suggests that any specialized detectors (e.g. corner-point detectors)can be approximated by very fast detection classifiers. In this paper, we analyzethe requirements on tools which are needed when experimenting with detection classifiersand we present a general framework which was created to fulfill theserequirements. This framework offers high performance for training, highvariability, elegant handling of configuration and it is able to meet all therequirements which arise when experimenting with almost all possible kinds ofdetection classifiers. The framework offers good testing support, fullsupporting infrastructure and some useful training algorithms and features. We offerthis framework for research and educational purposes and we hope it will allow lowerinitial investments when experimenting with detectionclassifiers.

Detection of patterns in images withclassifiers is currently one of the most important research topics in computervision. Many practical applications such as face detection exist and recentwork even suggests that any specialized detectors (e.g. corner-point detectors)can be approximated by very fast detection classifiers. In this paper, we analyzethe requirements on tools which are needed when experimenting with detection classifiersand we present a general framework which was created to fulfill theserequirements. This framework offers high performance for training, highvariability, elegant handling of configuration and it is able to meet all therequirements which arise when experimenting with almost all possible kinds ofdetection classifiers. The framework offers good testing support, fullsupporting infrastructure and some useful training algorithms and features. We offerthis framework for research and educational purposes and we hope it will allow lowerinitial investments when experimenting with detectionclassifiers.

Detection, Face Detection, Classification, Image Processing,Computer Vision, AdaBoost, WaldBoost, Cascade of Classifiers, Corner Points,Classifier Evaluation

Detection, Face Detection, Classification, Image Processing,Computer Vision, AdaBoost, WaldBoost, Cascade of Classifiers, Corner Points,Classifier Evaluation

HRADIŠ, M.

2010

22.04.2008

Comenius University in Bratislava

Budmerice

978-80-89186-30-3

Proceedings of Spring Conference on Computer Graphics

171

177

7

http://www.fit.vutbr.cz/research/groups/graph/publi/2008/2008-Hradis-SCCG-Framework.pdf

http://hdl.handle.net/