Improved repeatability measures for evaluating performance of feature detectors

TL;DR

This paper introduces improved repeatability measures for local feature detectors that better reflect true performance, revealing Hessian-based detectors as generally superior under various transformations.

Contribution

The paper proposes new repeatability formulations that more accurately correlate with actual detector performance, enhancing evaluation methods.

Findings

Hessian-based detectors outperform others under transformations

Improved measures show better correlation with true performance

Comparative analysis across state-of-the-art detectors

Abstract

The most frequently employed measure for performance characterisation of local feature detectors is repeatability, but it has been observed that this does not necessarily mirror actual performance. Presented are improved repeatability formulations which correlate much better with the true performance of feature detectors. Comparative results for several state-of-the-art feature detectors are presented using these measures; it is found that Hessian-based detectors are generally superior at identifying features when images are subject to various geometric and photometric transformations.