Multi-modal query expansion for video object instances retrieval

Abstract : In this paper we tackle the issue of object instances retrieval in video repositories using minimum information from the user (e.g., textual description/tags). Starting for a set of tags, images containing the object of interest are crawled from popular image search engines and repositories (e.g., Bing, Fickr, Google) and the positive and most representative instances of the object are automatically identified. These positive images are then used to generate a visual query descriptor and to retrieve videos containing the object of the interest. This multi-modal approach makes it possible to retrieve video content through images obtained from textual queries, without the use of any advanced learning technique. We test out method on the Flickr corpus of the TRECVID 2012 Instance Search Task.
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00944815
Contributor : Ruxandra Tapu <>
Submitted on : Tuesday, February 11, 2014 - 11:40:16 AM
Last modification on : Thursday, April 11, 2019 - 4:02:09 PM

Identifiers

  • HAL Id : hal-00944815, version 1

Citation

Andrei Bursuc, Zaharia Titus. Multi-modal query expansion for video object instances retrieval. MVA2013 IAPR International Conference on Machine Vision Applications, 2013, pp.214-217. ⟨hal-00944815⟩

Share

Metrics

Record views

452