Re-ranking for Multimedia Indexing and Retrieval

Bahjat Safadi (1) , Georges Quénot (2)

1 LIG - Laboratoire d'Informatique de Grenoble
2 MRIM - Modélisation et Recherche d’Information Multimédia [Grenoble]

We proposed a re-ranking method for improving the performance of semantic video indexing and retrieval. Experimental results show that the proposed re-ranking method is effective and it improves the system performance on average by about 16-22\% on TRECVID 2010 semantic indexing task.

Mots clés: Multimedia Indexing and Retrieval; Re-ranking

Domaines: Recherche d'information [cs.IR]

Bahjat Safadi, Georges Quénot. Re-ranking for Multimedia Indexing and Retrieval. ECIR 2011 - European Conference on IR Research, Apr 2011, Dublin, Ireland. pp.708-711, ⟨10.1007/978-3-642-20161-5_76⟩. ⟨hal-00763562⟩