Jump to : Download | Abstract | Contact | BibTex reference | EndNote reference |

Jiang:chapter01

Wei Jiang, Alexander Loui, Shih-Fu Chang. Cross-domain learning for semantic concept detection. In TV Content Analysis: Techniques and Applications (CRC Press), Taylor & Francis, 2011.

Download [help]

Download paper: Adobe portable document (pdf)

Copyright notice:This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Abstract

In this chapter, we investigate cross-domain learning methods that effectively incorporate information from available training resources in other domains to enhance concept detection in a target domain by considering the domain difference. Our contribution lies in two folds. First, we develop three approaches to incorporate three types of information to assist the target domain: the Cross-Domain Support Vector Machine (CDSVM) algorithm that uses previously learned support vectors; the prediction-based method that uses concept scores of the new data predicted by previously trained concept detectors; and the Adaptive Semi-Supervised SVM (AS3VM) algorithm that incrementally updates previously learned SVM concept detectors to classify new target data. Second, we provide a comprehensive summary and comparative study of the state-of-the-art SVM-based cross-domain learning methods

Contact

Wei Jiang
Shih-Fu Chang

BibTex Reference

@InCollection{Jiang:chapter01,
   Author = {Jiang, Wei and Loui, Alexander and Chang, Shih-Fu},
   Title = {Cross-domain learning for semantic concept detection},
   BookTitle = {TV Content Analysis: Techniques and Applications (CRC Press)},
   Publisher = {Taylor & Francis},
   Year = {2011}
}

EndNote Reference [help]

Get EndNote Reference (.ref)

 
bar

For problems or questions regarding this web site contact The Web Master.

This document was translated automatically from BibTEX by bib2html (Copyright 2003 © Eric Marchand, INRIA, Vista Project).