The Mediamill Semantic Video Search Engine

Author

Worring, M. ; Snoek, Cees G. M. ; de Rooij, Ork ; Nguyen, G.P. ; Smeulders, Arnold W. M.

Author_Institution

Intelligent Syst. Lab. Amsterdam, Amsterdam Univ., Netherlands

Volume

4

fYear

2007

fDate

15-20 April 2007

Abstract

In this paper we present the methods underlying the MediaMill semantic video search engine. The basis for the engine is a semantic indexing process which is currently based on a lexicon of 491 concept detectors. To support the user in navigating the collection, the system defines a visual similarity space, a semantic similarity space, a semantic thread space, and browsers to explore them. We compare the different browsers and their utility within the TRECVID benchmark. In 2005, we obtained a top-3 result for 19 out of 24 search topics. In 2006 for 14 out of 24.

Keywords

online front-ends; search engines; video signal processing; MediaMill semantic video search engine; TRECVID benchmark; concept detectors; semantic indexing process; semantic similarity; semantic thread; visual similarity; Data visualization; Detectors; Indexing; Intelligent systems; NIST; Navigation; Search engines; Speech; Statistics; Yarn; Video indexing; performance evaluation; retrieval; semantic threads; visualization;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference on

Conference_Location

Honolulu, HI

ISSN

1520-6149

Print_ISBN

1-4244-0727-3

Type

conf

DOI

10.1109/ICASSP.2007.367294

Filename

4218325