Rushes summarization by IRIM consortium: redundancy removal and multi-feature fusion

Georges Quénot (1, 2) , Jenny Benois-Pineau (3) , Boris Mansencal (3) , Eliana Rossi (3) , Matthieu Cord (4) , Frédéric Precioso (5) , David Gorisse (5) , Patrick Lambert (6) , Bertrand Augereau (7) , Lionel Granjon (8) , Denis Pellerin (8) , Michèle Rombaut (8) , Stéphane Ayache (9)

1 LIG - Laboratoire d'Informatique de Grenoble
2 MRIM - Modélisation et Recherche d’Information Multimédia [Grenoble]
3 LaBRI - Laboratoire Bordelais de Recherche en Informatique
4 MALIRE - Machine Learning and Information Retrieval
5 ETIS - UMR 8051 - Equipes Traitement de l'Information et Systèmes
6 LISTIC - Laboratoire d'Informatique, Systèmes, Traitement de l'Information et de la Connaissance
7 SIC
8 GIPSA-GPIG - GIPSA - Géométrie, Perception, Images, Geste
9 imagine - Extraction de Caractéristiques et Identification

In this paper, we present the first participation of a consortium of French laboratories, IRIM, to the TRECVID 2008 BBC Rushes Summarization task. Our approach resorts to video skimming. We propose two methods to reduce redundancy, as rushes include several takes of scenes. We also take into account low and midlevel semantic features in an ad-hoc fusion method in order to retain only significant content.

Mots clés: audio activity; information fusion.; information fusion; SBD; approximate k-NN clustering; HA clustering; motion features; mid-level-features; camera motion; face detection

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Georges Quénot, Jenny Benois-Pineau, Boris Mansencal, Eliana Rossi, Matthieu Cord, et al.. Rushes summarization by IRIM consortium: redundancy removal and multi-feature fusion. TRECVID BBC Rushes Summarization Workshop at ACM Multimedia'08, Oct 2008, Vancouver, BC, Canada. pp.80-84, ⟨10.1145/1463563.1463577⟩. ⟨hal-00347902⟩