Indexation de séquences de descripteurs

Romain Tavenard 1
1 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Getting information from multimedia documents is a very important field of research. We can now process huge image databases and do effective content-based searches. Processing more complex documents such as video or audio streams appears to be the next step in the development of content-based search tools. Video and audio streams are different as they embed the notion of descriptors sequences, in which the order between described elements is a key. This thesis proposes two indexing methods for temporal multimedia documents. The first one is based on the use of the Dynamic Time Warping (DTW) algorithm to compare sequences. This thesis introduces a method that presents significant improvement in terms of response time when compared to already existing methods. The second one is specifically applied to cover song detection. It consists in a first filtering stage of temporal regions of the database that are possible matches for elements of the query song and a second robustification stage that ensures temporal consistency.
Mots-clés : séquences indexation
Document type :
Theses
Liste complète des métadonnées

Cited literature [66 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-00639225
Contributor : Romain Tavenard <>
Submitted on : Tuesday, November 8, 2011 - 3:18:59 PM
Last modification on : Friday, November 16, 2018 - 1:23:31 AM
Document(s) archivé(s) le : Thursday, November 15, 2012 - 11:30:23 AM

Identifiers

  • HAL Id : tel-00639225, version 1

Citation

Romain Tavenard. Indexation de séquences de descripteurs. Multimédia [cs.MM]. Université Rennes 1, 2011. Français. ⟨tel-00639225⟩

Share

Metrics

Record views

464

Files downloads

569