International Journal of Computer Vision

, Volume 65, Issue 1, pp 43–72

A Comparison of Affine Region Detectors


    • University of Oxford
  • T. Tuytelaars
    • University of Leuven
  • C. Schmid
  • A. Zisserman
    • University of Oxford
  • J. Matas
    • Czech Technical University
  • F. Schaffalitzky
    • University of Oxford
  • T. Kadir
    • University of Oxford
  • L. Van Gool
    • University of Leuven

DOI: 10.1007/s11263-005-3848-x

Cite this article as:
Mikolajczyk, K., Tuytelaars, T., Schmid, C. et al. Int J Comput Vision (2005) 65: 43. doi:10.1007/s11263-005-3848-x


The paper gives a snapshot of the state of the art in affine covariant region detectors, and compares their performance on a set of test images under varying imaging conditions. Six types of detectors are included: detectors based on affine normalization around Harris  (Mikolajczyk and  Schmid, 2002; Schaffalitzky and  Zisserman, 2002) and Hessian points  (Mikolajczyk and  Schmid, 2002), a detector of ‘maximally stable extremal regions', proposed by Matas et al. (2002); an edge-based region detector  (Tuytelaars and Van Gool, 1999) and a detector based on intensity extrema (Tuytelaars and Van Gool, 2000), and a detector of ‘salient regions', proposed by Kadir, Zisserman and Brady (2004). The performance is measured against changes in viewpoint, scale, illumination, defocus and image compression.

The objective of this paper is also to establish a reference test set of images and performance software, so that future detectors can be evaluated in the same framework.


affine region detectorsinvariant image descriptionlocal featuresperformance evaluation

Copyright information

© Springer Science + Business Media, Inc. 2005