Evaluation metric for image understanding
- 1 November 2009
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 4381-4384
Abstract
We propose in this paper a new evaluation metric that enables to quantify the quality of an image interpretation result. This metric takes into account the a priori knowledge used by the interpretation algorithm and the ground truth associated with the original image. We combine two metrics that evaluate the localization and recognition results of each detected object. We show that the proposed metric fulfills some theoretical properties and has a correct behavior face to empirical experiments on an image benchmark database. We think that this metric could be a reliable reference for image and video understanding competitions.Keywords
This publication has 8 references indexed in Scilit:
- Evaluation Protocol for Localization Metrics2008
- Object count/area graphs for the evaluation of object detection and segmentation algorithmsInternational Journal on Document Analysis and Recognition (IJDAR), 2006
- Histograms of Oriented Gradients for Human Detection2005
- Scale-invariant shape features for recognition of object categories2004
- Detecting moving objects, ghosts, and shadows in video streamsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2003
- A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics2002
- Empirical performance evaluation of graphics recognition systemsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1999
- Algorithms for the Assignment and Transportation ProblemsJournal of the Society for Industrial and Applied Mathematics, 1957