Topic Tracking Across Broadcast News Videos with Visual Duplicates and Semantic Concepts
- 1 October 2006
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 141-144
- https://doi.org/10.1109/icip.2006.312379
Abstract
Videos from distributed sources (e.g., broadcasts, podcasts, blogs, etc.) have grown exponentially. Topic threading is very useful for organizing such large-volume information sources. Current solutions primarily rely on text features only but encounter difficulty when text is noisy or unavailable. In this paper, we propose new representations and similarity measures for news videos based on low-level features, visual near-duplicates, and high-level semantic concepts automatically detected from videos. We develop a multi-modal fusion framework for estimating relevance of a new story to a known topic. Our extensive experiments using TRECVID 2005 data set (171 hours, 6 channels, 3 languages) confirm that near-duplicates consistently and significantly boost the tracking performance by up to 25%. In addition, we present information-theoretic analysis to assess the complexity of each semantic topic and determine the best subset of concepts for tracking each topic.Keywords
This publication has 5 references indexed in Scilit:
- Tracking news stories across different sourcesPublished by Association for Computing Machinery (ACM) ,2005
- Visual Concepts for News Story Tracking: Analyzing and Exploiting the NIST TRECVID Video Annotation ExperimentPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Detecting image near-duplicate by stochastic attributed relational graph matching with learningPublished by Association for Computing Machinery (ACM) ,2004
- Document clustering using word clusters via the information bottleneck methodPublished by Association for Computing Machinery (ACM) ,2000
- Learning approaches for detecting and tracking news eventsIEEE Intelligent Systems and their Applications, 1999