Evaluation of interest point detectors

Citation
C. Schmid et al., Evaluation of interest point detectors, INT J COM V, 37(2), 2000, pp. 151-172
Citations number
49
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
INTERNATIONAL JOURNAL OF COMPUTER VISION
ISSN journal
09205691 → ACNP
Volume
37
Issue
2
Year of publication
2000
Pages
151 - 172
Database
ISI
SICI code
0920-5691(200006)37:2<151:EOIPD>2.0.ZU;2-X
Abstract
Many different low-level feature detectors exist and it is widely agreed th at the evaluation of detectors is important. In this paper we introduce two evaluation criteria for interest points: repeatability rate and informatio n content. Repeatability rate evaluates the geometric stability under diffe rent transformations. Information content measures the distinctiveness of f eatures. Different interest point detectors are compared using these two cr iteria. We determine which detector gives the best results and show that it satisfies the criteria well.