International Journal of Computer Vision

, Volume 108, Issue 1–2, pp 115–132

Iterative Category Discovery via Multiple Kernel Metric Learning

  • Carolina Galleguillos
  • Brian McFee
  • Gert R. G. Lanckriet
Article

DOI: 10.1007/s11263-013-0679-z

Cite this article as:
Galleguillos, C., McFee, B. & Lanckriet, G.R.G. Int J Comput Vis (2014) 108: 115. doi:10.1007/s11263-013-0679-z

Abstract

The goal of an object category discovery system is to annotate a pool of unlabeled image data, where the set of labels is initially unknown to the system, and must therefore be discovered over time by querying a human annotator. The annotated data is then used to train object detectors in a standard supervised learning setting, possibly in conjunction with category discovery itself. Category discovery systems can be evaluated in terms of both accuracy of the resulting object detectors, and the efficiency with which they discover categories and annotate the training data. To improve the accuracy and efficiency of category discovery, we propose an iterative framework which alternates between optimizing nearest neighbor classification for known categories with multiple kernel metric learning, and detecting clusters of unlabeled image regions likely to belong to a novel, unknown categories. Experimental results on the MSRC and PASCAL VOC2007 data sets show that the proposed method improves clustering for category discovery, and efficiently annotates image regions belonging to the discovered classes.

Keywords

Category discovery Metric learning  Multiple kernel learning Iterative discovery 

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Carolina Galleguillos
    • 1
  • Brian McFee
    • 2
  • Gert R. G. Lanckriet
    • 3
  1. 1.SET Media IncSan FranciscoUSA
  2. 2.Columbia UniversityNew YorkUSA
  3. 3.University of California, San DiegoLa JollaUSA

Personalised recommendations