Communication Dans Un Congrès Année : 2013

Robust TV Stream Labelling with Conditional Random Fields

Résumé

Multi-label video annotation is a challenging task and a necessary first step for further processing. In this paper, we investigate the task of labelling TV stream segments into programs or several types of breaks through machine learning. Our contribution is twofold: 1) we propose to use simple yet efficient descriptors for this labelling task, 2) we show that Conditional Random Fields (CRF) are especially suited for this task. In particular, through several experiments, we show that CRF out-perform other machine learning techniques, while requiring few training data thanks to its ability to handle the different types of sequential information lying in our data.

Fichier principal
Vignette du fichier
Ncibi-Mmedia2012.pdf (206.55 Ko) Télécharger le fichier
Loading...

Dates et versions

hal-00844640 , version 1 (20-10-2016)

Licence

Identifiants

  • HAL Id : hal-00844640 , version 1

Citer

Abir Ncibi, Emmanuelle Martienne, Vincent Claveau, Guillaume Gravier, Patrick Gros. Robust TV Stream Labelling with Conditional Random Fields. MMEDIA - 5th International Conference on Advances in Multimedia, IARIA, Apr 2013, Venise, Italy. pp.88-95. ⟨hal-00844640⟩
692 Consultations
197 Téléchargements

Partager

  • More