International Journal of Computer Vision

, Volume 115, Issue 1, pp 29–43

Predicting Entry-Level Categories

  • Vicente Ordonez
  • Wei Liu
  • Jia Deng
  • Yejin Choi
  • Alexander C. Berg
  • Tamara L. Berg
Article

DOI: 10.1007/s11263-015-0815-z

Cite this article as:
Ordonez, V., Liu, W., Deng, J. et al. Int J Comput Vis (2015) 115: 29. doi:10.1007/s11263-015-0815-z

Abstract

Entry-level categories—the labels people use to name an object—were originally defined and studied by psychologists in the 1970s and 1980s. In this paper we extend these ideas to study entry-level categories at a larger scale and to learn models that can automatically predict entry-level categories for images. Our models combine visual recognition predictions with linguistic resources like WordNet and proxies for word “naturalness” mined from the enormous amount of text on the web. We demonstrate the usefulness of our models for predicting nouns (entry-level words) associated with images by people, and for learning mappings between concepts predicted by existing visual recognition systems and entry-level concepts. In this work we make use of recent successful efforts on convolutional network models for visual recognition by training classifiers for 7404 object categories on ConvNet activation features. Results for category mapping and entry-level category prediction for images show promise for producing more natural human-like labels. We also demonstrate the potential applicability of our results to the task of image description generation.

Keywords

Recognition Categorization Entry-level categories Psychology 

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  • Vicente Ordonez
    • 1
  • Wei Liu
    • 1
  • Jia Deng
    • 2
  • Yejin Choi
    • 3
  • Alexander C. Berg
    • 1
  • Tamara L. Berg
    • 1
  1. 1.Department of Computer ScienceUniversity of North Carolina at Chapel HillChapel HillUSA
  2. 2.Department of Electrical Engineering and Computer ScienceUniversity of MichiganAnn ArborUSA
  3. 3.Computer Science and EngineeringUniversity of WashingtonSeattleUSA