Generative versus discriminative methods for object recognition

I. Ulusoy, C.M. Bishop

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Many approaches to object recognition are founded on probability theory, and can be broadly characterized as either generative or discriminative according to whether or not the distribution of the image features is modelled. Generative and discriminative methods have very different characteristics, as well as complementary strengths and weaknesses. In this paper we introduce new generative and discriminative models for object detection and classification based on weakly labelled training data. We use these models to illustrate the relative merits of the two approaches in the context of a data set of widely varying images of non-rigid objects (animals). Our results support the assertion that neither approach alone will be sufficient for large scale object recognition, and we discuss techniques for combining them.
Original languageEnglish
Title of host publicationComputer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on
PublisherIEE
Pages258-265 vol. 2
Volume2
ISBN (Print)0-7695-2372-2
DOIs
Publication statusPublished - 1 Jun 2005

Keywords

  • feature extraction
  • image classification
  • learning (artificial intelligence)
  • object recognition
  • probability
  • discriminative method
  • generative method
  • image feature
  • object classification
  • probability theory
  • weakly labelled training data
  • Animals
  • Character generation
  • Computer vision
  • Context modeling
  • Large-scale systems
  • Machine learning
  • Object detection
  • Object recognition
  • Predictive models
  • Training data

Fingerprint Dive into the research topics of 'Generative versus discriminative methods for object recognition'. Together they form a unique fingerprint.

Cite this