Abstract : In this paper we present the results of our experiments in the Trecvid'10 copy detection task and introduce the components of our system. In particular, we describe the recent approximate nearest neighbor search method we used to index the hundreds millions of audio descriptors. Our system obtained excellent accuracy and localization results, achieving the best performance on a few transformations, and this with a single kind of image descriptor. Moreover, the analysis evidences that our system can be significantly improved.
https://hal.inria.fr/inria-00555825
Contributor : Hervé Jégou
<>
Submitted on : Wednesday, February 9, 2011 - 11:46:24 AM
Last modification on : Thursday, February 7, 2019 - 2:35:52 PM
Document(s) archivé(s) le : Tuesday, November 6, 2012 - 1:41:00 PM
Hervé Jégou, Matthijs Douze, Guillaume Gravier, Cordelia Schmid, Patrick Gros. INRIA LEAR-TEXMEX: Video copy detection task. TRECVid 2010 Workshop, Nov 2010, Gaithersburg, United States. ⟨inria-00555825⟩
