meta data for this page

Gabor Features in Computer Vision and Image Processing

I have used Gabor features in various computer vision and image processing tasks for about ten years now and it certainly is my favorite topic. Every now and then I have been asked to give an introductory lecture to the topic and finally in 2010 I decided to collect all relevant information to this wiki-page for the future.

- Joni Kamarainen (May, 2010)


2D Gabor filters have maintained their popularity as a multi-purpose feature extraction method in computer vision and image processing for almost three decades. The original article of Nobel laureate Dennis Gabor dates back to 1946, but the most influential work in these fields is John Daugman's article in 1985.

During the 00's the research activity using Gabor features has again increased according to IEEE Xplore database. The most important reason is their remarkable success in emerging application areas, such as biometric authentication. Daugman's Iris Code is The Method for iris recognition, methods based on Gabor features produce top scores in face recognition challenges (e.g. the two best in the ICPR 2004 contest), and provide state-of-the-art accuracy in fingerprint matching (Jain et al. PAMI 2007) and face detection (Hamouz et al. PAMI 2005). Even the old adage of correspondence to simple cells in the visual cortex has recently been resurrected (Serre et al. PAMI 2007). Since their invention, Gabor features have succeeded in numerous applications, and therefore, are important and general tools every computer vision and image processing scientist should know.


The current tutorial material will take from three to four hours including breaks. The tutorial is divided to three different topics:

1 Fundamentals (Refs. [1],[2] and [3])

  • Background
  • Principles - “Theory of Communication”
  • Gabor features

2 Usage (Refs. [2], [4], [5])

  • Fundamentals
  • Feature Construction
  • Multi-resolution Gabor feature - ``Simple Gabor feature space''

3 Applications (Refs. [6], [7], [8]) (due to copyright issues ask handout by email)

  • Usage of Gabor Filters
  • Face recognition - Gabor jets and elastic bunch graph matching
  • Iris recognition - Daugman's phase descriptor
  • Supervised object detection - Multi-resolution Gabor features

In the first part, we will go through the main idea behind Gabor function and specifically follow the idea in the Gabor's original study [1]. You will learn the importance of the two domains - time and frequency - for analyzing signals. In the second part, we learn how to construct features from the Gabor filter responses and the main properties of the features. In the last part, we will walk through the most important and influential works using Gabor features. The three presented applications utilize the filters very differently.

Unfortunately I cannot make the slides publicly available since they contain copyrighted material. All participants will however have the printed handouts.



  1. [1] D. Gabor, Theory of Communication, Journal of Institution of Electrical Engineers (93) 1946, 429-457. PDF
  2. [2] J.-K. Kamarainen, Feature Extraction Using Gabor Filters, PhD thesis, Lappeenranta University of Technology, 2003. PDF
  3. [3] J.G. Daugman, Uncertainty relation for resolution in space, spatial frequency, and orientation optimized by two-dimensional visual cortical filters, Journal of Optical Society of America A 2, 7 (1985). PDF
  4. [4] Kamarainen, J.-K., Kyrki, V., Kälviäinen, H., Invariance Properties of Gabor Filter Based Features - Overview and Applications, IEEE Transactions on Image Processing 15, 5 (2006) 1088-1099.
  5. [5] Ilonen, J., Kamarainen, J.-K., Paalanen, P., Hamouz, M., Kittler, J., Kälviäinen, H., Image feature localization by multiple hypothesis testing of Gabor features, IEEE Transactions on Image Processing 17, 3 (2008) 311-325.
  6. [6] L. Wiskott, J.-M. Fellous, N. Krueger, C. von der Malsburg, Face Recognition by Elastic Bunch Graph Matching, IEEE Transactions on Pattern Analysis and Machine Intelligence 19, 7 (1997). PDF More recent thesis (PDF)
  7. [7] J.G. Daugman, High Confidence Visual Recognition of Persons by a Test of Statistical Independence, IEEE Transactions on Pattern Analysis and Machine Intelligence 15, 11 (1993). PDF
  8. [8] Kamarainen, J.-K., Hamouz, M., Kittler, J., Paalanen, P., Ilonen, J., Drobchenko, A., Object Localisation Using Generative Probability Model for Spatial Constellation and Local Image Features, ICCV 2007 Workshop on Non-Rigid Registration and Tracking Through Learning (NRTL2007) (Rio de Janeiro, Brazil, 2007). PDF