Automatic extraction of story units from TV news

Bogdan Mocanu (1, 2, 3, 4, 5) , Ruxandra Tapu (1, 2, 3, 4, 5) , Titus Zaharia (6, 1, 2, 3, 5)

1 TSP - ARTEMIS - Département Advanced Research And Techniques For Multidimensional Imaging Systems
2 SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
3 MAP5 - UMR 8145 - Mathématiques Appliquées Paris 5
4 UPB - University Politehnica of Bucharest [Romania]
5 CNRS - Centre National de la Recherche Scientifique
6 ARMEDIA-SAMOVAR - ARMEDIA

In this paper we propose a novel method for automatic identification of semantically consistent story units from TV news programs. The method includes a temporal segmentation procedure of the video based on visual cues, as well as a graph-driven textual analysis technique of the subtitles documents. The experimental results, obtained on a dataset of 50 videos selected from one week video archive of France Television demonstrate the pertinence of the proposed approach

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

Bogdan Mocanu, Ruxandra Tapu, Titus Zaharia. Automatic extraction of story units from TV news. ICCE 2017 : International Conference on Consumer Electronics, Jan 2017, Las Vegas United States. pp.1 - 2, ⟨10.1109/ICCE.2017.7889377⟩. ⟨hal-01511894⟩