Network Visual Analytics of News Videos based on Facetracking
Résumé
The current advances in the domain of computer vision bring us new tools to perform high-level analysis. News analysis has been an important focus of visual analytics systems. Social networks analysis is also a key tool for conducting sociologic studies in news media. In this paper, we propose to bring the three domains together: we present a visual analytics system to explore news media from social networks derived from video analysis. The social networks are created from facial detection and tracking of over ten years of the Japanese NHK news archive. The interactive visual exploration of the different social networks offers then a comparative view of the politico-media scene presented by NHK between 2001 and 2013.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...