Evaluation Protocol for Localization Metrics

Authors:
Baptiste Hemery;Hélène Laurent;Christophe Rosenberger;Bruno Emile
Affiliations:
Institut Prisme ENSI de Bourges, Université d'Orléans, Bourges, France 18020;Institut Prisme ENSI de Bourges, Université d'Orléans, Bourges, France 18020;Laboratoire Greyc ENSICAEN, Université de Caen - CNRS, Caen, France 14000;Institut Prisme ENSI de Bourges, Université d'Orléans, Bourges, France 18020
Venue:
ICISP '08 Proceedings of the 3rd international conference on Image and Signal Processing
Year:
2008

Citing 4
Cited 0

Distance measures for signal processing and pattern recognition

Signal Processing
A New Metric for Grey-Scale Image Comparison

International Journal of Computer Vision
Quantitative methods of evaluating image segmentation

ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3
A new supervised evaluation criterion for region based segmentation methods

ACIVS'07 Proceedings of the 9th international conference on Advanced concepts for intelligent vision systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Localization metrics permit to quantify the correctness of object detection in an image interpretation result. This paper deals with the definition of a protocol in order to evaluate the behavior of localization metrics. We first define some properties that metrics should verify and create a synthetic database that enables to verify those properties on different metrics. After presenting the tested localization metrics, the results obtained following the proposed protocol are exposed. Finally, some conclusions and perspectives are given.