Hierarchically Multi-modal Indexing of Sports Video

Author

Liu, Yuchi ; Wu, Lingda ; Xie, Yuxiang ; Lei, Zhen ; Jian, Jie

Author_Institution

Sch. of Inf. Syst. & Manage., Nat. Univ. of Defense Technol., Changsha

fYear

0

fDate

0-0 0

Firstpage

533

Lastpage

537

Abstract

A multi-level structure of video contents is introduced and a novel mechanism of hierarchically indexing sports video using multi-modal cues is proposed. Firstly, video is indexed with information from video and audio individually. Video is segmented into physical shots based on visual features, and then identified as syntactic shots on the middle level features according to broadcasting characteristics. Audio is analyzed to get physical and syntactic contents step by step. Then, fusing the informative clues from video, audio and text extracted from video, semantics of sports video can be generated and shots are annotated as semantic shots. Considering the difficulty of extracting full semantics of video, a multi-modal retrieval is proposed to compensate query by event

Keywords

audio signal processing; feature extraction; image segmentation; indexing; video retrieval; audio analysis; hierarchical multimodal indexing; multimodal retrieval; sports video; text extraction; visual feature; Broadcasting; Content based retrieval; Data mining; Games; Indexing; Information retrieval; Management information systems; Multimedia communication; Speech analysis; Streaming media; multi-level structure of video contents; multi-modal indexing; multi-modal retrieval; sports video; syntactic contents;

fLanguage

English

Publisher

ieee

Conference_Titel

Information, Communications and Signal Processing, 2005 Fifth International Conference on

Conference_Location

Bangkok

Print_ISBN

0-7803-9283-3

Type

conf

DOI

10.1109/ICICS.2005.1689103

Filename

1689103