Image Feature Localization by Multiple Hypothesis Testing of Gabor Features

Jarmo Ilonen, Joni-Kristian Kamarainen, Pekka Paalanen, Miroslav Hamouz, Josef Kittler, Heikki KälviäinenIEEE TIP 2008 · DOI · OpenReview

Abstract

Several novel and particularly successful object and object category detection and recognition methods based on image features, local descriptions of object appearance, have recently been proposed. The methods are based on a localization of image features and a spatial constellation search over the localized features. The accuracy and reliability of the methods depend on the success of both tasks: image feature localization and spatial constellation model search. In this paper, we present an improved algorithm for image feature localization. The method is based on complex-valued multiresolution Gabor features and their ranking using multiple hypothesis testing. The algorithm provides very accurate local image features over arbitrary scale and rotation. We discuss in detail issues such as selection of filter parameters, confidence measure, and the magnitude versus complex representation, and show on a large test sample how these influence the performance. The versatility and accuracy of the method is demonstrated on two profoundly different challenging problems (faces and license plates).
BibTeX
@article{ilonen2008image,
  author = {Jarmo Ilonen and Joni-Kristian Kamarainen and Pekka Paalanen and Miroslav Hamouz and Josef Kittler and Heikki Kälviäinen},
  title = {{Image Feature Localization by Multiple Hypothesis Testing of Gabor Features}},
  journal = {IEEE Trans. Image Process.},
  year = {2008},
  doi = {10.1109/tip.2007.916052},
}

Loading related content...

Figure