Focused evaluation for image description with binary forced-choice tasks

Micah Hodosh, Julia Hockenmaier

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

Abstract

Current evaluation metrics for image description may be too coarse. We therefore propose a series of binary forced-choice tasks that each focus on a different aspect of the captions. We evaluate a number of different off-the-shelf image description systems. Our results indicate strengths and shortcomings of both generation and ranking based approaches.

Original languageEnglish (US)
Title of host publicationProceedings of the 5th Workshop on Vision and Language, VL 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016
EditorsAnya Belz, Erkut Erdem, Krystian Mikolajczyk, Katerina Pastra
PublisherAssociation for Computational Linguistics (ACL)
Pages19-28
Number of pages10
ISBN (Electronic)9781945626111
StatePublished - 2016
Event5th Workshop on Vision and Language, VL 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016 - Berlin, Germany
Duration: Aug 12 2016 → …

Publication series

NameProceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN (Print)0736-587X

Conference

Conference5th Workshop on Vision and Language, VL 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016
Country/TerritoryGermany
CityBerlin
Period8/12/16 → …

ASJC Scopus subject areas

  • Computer Science Applications
  • Linguistics and Language
  • Language and Linguistics

Fingerprint

Dive into the research topics of 'Focused evaluation for image description with binary forced-choice tasks'. Together they form a unique fingerprint.

Cite this