Classification with global, local and shared features

Hakan Bilen, Vinay P. Namboodiri, Luc J. Van Gool

Research output: Chapter or section in a book/report/conference proceedingChapter in a published conference proceeding

4 Citations (SciVal)

Abstract

We present a framework that jointly learns and then uses multiple image windows for improved classification. Apart from using the entire image content as context, class-specific windows are added, as well as windows that target class pairs. The location and extent of the windows are set automatically by handling the window parameters as latent variables. This framework makes the following contributions: a) the addition of localized information through the class-specific windows improves classification, b) windows introduced for the classification of class pairs further improve the results, c) the windows and classification parameters can be effectively learnt using a discriminative max-margin approach with latent variables, and d) the same framework is suited for multiple visual tasks such as classifying objects, scenes and actions. Experiments demonstrate the aforementioned claims.

Original languageEnglish
Title of host publicationPattern Recognition - Joint 34th DAGM and 36th OAGM Symposium, Proceedings
Pages134-143
Number of pages10
DOIs
Publication statusPublished - 11 Sept 2012
EventJoint 34th Symposium of the German Association for Pattern Recognition, DAGM 2012 and 36th Symposium of the Austrian Association for Pattern Recognition, OAGM 2012 - Graz, Austria
Duration: 28 Aug 201231 Aug 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7476 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

ConferenceJoint 34th Symposium of the German Association for Pattern Recognition, DAGM 2012 and 36th Symposium of the Austrian Association for Pattern Recognition, OAGM 2012
Country/TerritoryAustria
CityGraz
Period28/08/1231/08/12

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Classification with global, local and shared features'. Together they form a unique fingerprint.

Cite this