Database Lecture Series Talk - Dr. Ewa Kijak - Audiovisual Integration for Tennis Broadcast Structuring

On Tuesday, December 16th, at 12:00-13:00,  Dr. Ewa Kijak (University of Rennes 1, France) will give a Database Lecture Series Talk entitled "Audiovisual Integration for Tennis Broadcast Structuring" The lecture will be held in K-5 (Kringlan 1).

This talk will focus on the integration of multimodal features for video indexing. Sport video structure analysis is taken as applicative example. The method relies on a statistical model which takes into account both the shot content and the interleaving of shots. This stochastic modelling is performed in the global framework of Hidden Markov Models (HMMs) that can be efficiently applied to merge audio and visual cues. Our approach is validated in the particular domain of tennis videos. The model integrates prior information about tennis content and editing rules. The basic temporal unit is the video shot. Visual features are used to characterize the type of shot view.  Audio features describe the audio events within a video shot. As a result, typical tennis scenes are simultaneously segmented and identified.

Dr. Ewa Kijak is assistant professor at University of Rennes 1, France, working at the IRISA lab in the TEXMEX group. Her research interests are in the fields of images and videos description and indexing. She worked at Thomson Corporate Research, and then has been an assistant professor at University of Paris 6, working in the Computer Science Laboratory (LIP6) of the university during 2 years. She worked on video indexing and multimedia information fusion.