A Hybrid Approach to Improving Semantic Extraction of News Video
Authors:
- 1 September 2007
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 79-86
Abstract
In this paper we describe a hybrid approach to improving semantic extraction from news video. Experiments show the value of careful parameter tuning, exploiting multiple feature sets and multilingual linguistic resources, applying text retrieval approaches for image features, and establishing synergy between multiple concepts through undirected graphical models. No single approach provides a consistently better result for every concept detection, which suggests that extracting video semantics should exploit multiple resources and techniques rather than a single approach.Keywords
This publication has 12 references indexed in Scilit:
- Can High-Level Concepts Fill the Semantic Gap in Video Retrieval? A Case Study With Broadcast NewsIEEE Transactions on Multimedia, 2007
- Learning the semantics of multimedia queries and concepts from a small number of examples2005
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004
- Optimal multimodal fusion for multimedia data analysis2004
- Interactive content-based retrieval of video2003
- Automatic image annotation and retrieval using cross-media relevance models2003
- News video classification using SVM-based multimodal classifiers and combination strategies2002
- Does organisation by similarity assist image browsing?2001
- Content-based image retrieval at the end of the early yearsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- End-User Searching Challenges Indexing Practices in the Digital Newspaper Photo ArchiveInformation Retrieval Journal, 2000