Audio-visual content-based violent scene characterization

We present a novel technique to characterize and index violent scenes in general TV drama and movies. Our goal is to identify violent signatures and localize violent events within a movie to support "high-level" video indexing. In particular, we exploit multiple "audio-visual" signatures to create a perceptual relation for conceptually meaningful violent scene identification. Potential applications are automatic blocking of violence in movies watched by children, hiding violence using data hiding or information filtering and genre classification of digital video database.