Search of objects of interest in videos

Author

Mansencal, Boris ; Benois-Pineau, Jenny ; Vieux, Remi ; Domenger, Jean-Philippe

Author_Institution

LaBRI, Univ. Bordeaux, Talence, France

fYear

2012

fDate

27-29 June 2012

Firstpage

1

Lastpage

6

Abstract

The paper addresses the problem of object search in video content. Both Query-By-Example paradigm and context search are explored. In QBE paradigm the object of interest is searched by matching of object signatures built from SURF descriptors with on-the-fly computed signatures in frames. The ”context” search is understood as a query on the whole frame with features extracted after a region-based segmentation. Both kinds of features are transcribed in Bag-Of-Words framework. The combination of Bag-of-Visual-Words and Bag-of-Region-Words gives promising results in TRECVID´2011 Instance Search Task.

Keywords

feature extraction; image matching; image segmentation; video retrieval; video signal processing; SURF descriptor; TRECVID´2011 instance search task; bag-of-region-word; bag-of-words framework; context search; features extraction; object search problem; object signature matching; on-the-fly computed signature; query-by-example paradigm; region-based segmentation; video content; Context; Databases; Feature extraction; Image segmentation; Search problems; Videos; Visualization;

fLanguage

English

Publisher

ieee

Conference_Titel

Content-Based Multimedia Indexing (CBMI), 2012 10th International Workshop on

Conference_Location

Annecy

ISSN

1949-3983

Print_ISBN

978-1-4673-2368-0

Electronic_ISBN

1949-3983

Type

conf

DOI

10.1109/CBMI.2012.6269809

Filename

6269809