Bag-of-Features Codebook Generation by Self-Organisation

  • Authors:

  • Teemu Kinnunen;Joni-Kristian Kamarainen;Lasse Lensu;Heikki Kälviäinen

  • Affiliations:

  • Machine Vision and Pattern Recognition Laboratory (MVPR), MVPR/Computational Vision Group, Kouvola Department of Information Technology, Lappeenranta University of Technology, Finland;Machine Vision and Pattern Recognition Laboratory (MVPR), MVPR/Computational Vision Group, Kouvola Department of Information Technology, Lappeenranta University of Technology, Finland;Machine Vision and Pattern Recognition Laboratory (MVPR), MVPR/Computational Vision Group, Kouvola Department of Information Technology, Lappeenranta University of Technology, Finland;Machine Vision and Pattern Recognition Laboratory (MVPR), MVPR/Computational Vision Group, Kouvola Department of Information Technology, Lappeenranta University of Technology, Finland

  • Venue:
  • WSOM '09 Proceedings of the 7th International Workshop on Advances in Self-Organizing Maps
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Bag of features is a well established technique for the visual categorisation of objects, categories of objects and textures. One of the most important part of this technique is codebook generation since its within-class and between-class discrimination power is the main factor in the categorisation accuracy. A codebook is generated from regions of interest extracted automatically from a set of labeled (supervised/semi-supervised) or unlabeled (unsupervised) images. A standard tool for the codebook generation is the c-means clustering algorithm, and the state-of-the-art results have been reported using generation schemes based on the c-means. In this work, we challenge this mainstream approach by demonstrating how the competitive learning principle in the self-organising map (SOM) is able to provide similar and often superior results to the c-means. Therefore, we claim that exploiting the self-organisation principle is an alternative research direction to the mainstream research in visual object categorisation and its importance for the ultimate challenge, unsupervised visual object categorisation, needs to be investigated.