News video story segmentation using fusion of multi-level multi-modal features in TRECVID 2003

W. Hsu, L. Kennedy, C. W. Huang, S. F. Chang, C. Y. Lin, G. Iyengar

Research output: Contribution to journalConference articlepeer-review

25 Scopus citations

Abstract

In this paper, we present our new results in news video story segmentation and classification in the context of TRECVID video retrieval benchmarking event 2003. We applied and extended the Maximum Entropy statistical model to effectively fuse diverse features from multiple levels and modalities, including visual, audio, and text. We have included various features such as motion, face, music/speech types, prosody, and high-level text segmentation information. The statistical fusion model is used to automatically discover relevant features contributing to the detection of story boundaries. One novel aspect of our method is the use of a feature wrapper to address different types of features - asynchronous, discrete, continuous and delta ones. We also developed several novel features related to prosody. Using the large news video set from the TRECVID 2003 benchmark, we demonstrate satisfactory performance (F1 measure up to 0.76 ) and more importantly observe an interesting opportunity for further improvement.

Original languageEnglish
Pages (from-to)III645-III648
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume3
StatePublished - 2004
EventProceedings - IEEE International Conference on Acoustics, Speech, and Signal Processing - Montreal, Que, Canada
Duration: 17 May 200421 May 2004

Fingerprint

Dive into the research topics of 'News video story segmentation using fusion of multi-level multi-modal features in TRECVID 2003'. Together they form a unique fingerprint.

Cite this