Learning non-redundant codebooks for classifying complex objects

  • Authors:

  • Wei Zhang;Akshat Surve;Xiaoli Fern;Thomas Dietterich

  • Affiliations:

  • Hewlett-Packard Laboratories, Palo Alto, California, United States;Oregon State University, Corvallis, Oregon, United States;Oregon State University, Corvallis, Oregon, United States;Oregon State University, Corvallis, Oregon, United States

  • Venue:
  • ICML '09 Proceedings of the 26th Annual International Conference on Machine Learning
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Codebook-based representations are widely employed in the classification of complex objects such as images and documents. Most previous codebook-based methods construct a single codebook via clustering that maps a bag of low-level features into a fixed-length histogram that describes the distribution of these features. This paper describes a simple yet effective framework for learning multiple non-redundant codebooks that produces surprisingly good results. In this framework, each codebook is learned in sequence to extract discriminative information that was not captured by preceding codebooks and their corresponding classifiers. We apply this framework to two application domains: visual object categorization and document classification. Experiments on large classification tasks show substantial improvements in performance compared to a single codebook or codebooks learned in a bagging style.