Visual Classifier Prediction by Distributional Semantic Embedding of Text Descriptions

Mohamed Elhoseiny, Ahmed Elgammal

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

One of the main challenges for scaling up object recognition systems is the lack of annotated images for real-world categories. It is estimated that humans can recognize and discriminate among about 30,000 categories (Biederman and others, 1987). Typically there are few images available for training classifiers form most of these categories. This is reflected in the number of images per category available for training in most object categorization datasets, which, as pointed out in (Salakhutdinov et al., 2011), shows a Zipf distribution.

Original languageEnglish (US)
Title of host publicationA Workshop of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015 - Workshop on Vision and Language 2015, VL 2015
Subtitle of host publicationVision and Language Meet Cognitive Systems - Proceedings
PublisherAssociation for Computational Linguistics (ACL)
Pages48-50
Number of pages3
ISBN (Electronic)9781941643327
StatePublished - 2015
Event4th Workshop on Vision and Language, VL 2015, as part of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015 - Lisbon, Portugal
Duration: Sep 18 2015 → …

Publication series

NameA Workshop of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015 - Workshop on Vision and Language 2015, VL 2015: Vision and Language Meet Cognitive Systems - Proceedings

Conference

Conference4th Workshop on Vision and Language, VL 2015, as part of the 2015 Conference on Empirical Methods in Natural Language Processing, EMNLP 2015
Country/TerritoryPortugal
CityLisbon
Period09/18/15 → …

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Computational Theory and Mathematics

Fingerprint

Dive into the research topics of 'Visual Classifier Prediction by Distributional Semantic Embedding of Text Descriptions'. Together they form a unique fingerprint.

Cite this