Learning to Identify TV News Monologues by Style and Context

Abstract

This research focused on the problem of learning semantics from multimedia data associated with broadcast video documents. The authors proposed to learn semantic concepts from multimodal sources based on style and context detectors, in combination with statistical classifier ensembles. As a case study, they present their method for detecting the concept of news subject monologues. This approach had the best average precision performance amongst 26 submissions in the 2003 video track of the Text Retrieval Conference benchmark. Experiments were conducted with respect to individual detector contribution, ensemble size, and ranking mechanism. It was found that the combination of detectors is decisive for the final result, although some detectors might appear useless in isolation. Moreover, by using a probabilistic ranking in combination with a large classifier ensemble, the results can be improved even further.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 2003
Accession Number
ADA458538

Entities

People

  • Alexander G. Hauptmann
  • Cees G. Snoek

Organizations

  • Carnegie Mellon University

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Abstracts
  • Artificial Intelligence Software
  • Bayesian Networks
  • Character Recognition
  • Computer Science
  • Computers
  • Detection
  • Detectors
  • Information Science
  • Learning
  • Machine Learning
  • Pattern Recognition
  • Probability
  • Recognition
  • Supervised Machine Learning
  • Test And Evaluation
  • Test Sets

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Image Processing and Computer Vision.
  • Regression Analysis.