A New Evaluation Framework and Image Dataset for Keypoint Extraction and Feature Descriptor Matching

@Article {
author = {Iñigo Barandiaran, Camilo Cortés, Marcos Nieto, Manuel Graña, Oscar Ruiz},
title = {A New Evaluation Framework and Image Dataset for Keypoint Extraction and Feature Descriptor Matching},
pages = {252-257},
volume = {1},
keywds = {

Interest Points, Descriptors, Computer Vision, Data Set

}
abstract = {

Key point extraction and description mechanisms play a crucial role for image matching, where several image points must be accurately identified to robustly estimate a transformation or recognize an object or a scene. Currently several new mechanisms for key point extraction and for feature description are emerging, so normalized data and evaluation protocols are needed in order to assess them accurately. In response to these needs, we present a new evaluation framework for measuring different aspects and behaviours of the state-of-the-art feature point extraction and description mechanisms. In addition, we also propose a new image dataset and a testing image generator. This evaluation framework and dataset can be useful to help the research community improving their key point extraction and feature descriptor approaches. Also, the practitioners on computer vision applications, based on image point matching, can obtain valuable information from this contribution to select the algorithm that best suit their needs. All proposed material in this work is freely available on-line.

}
isi = {1},
date = {2013-02-21},
year = {2013},
}