No abstract

This paper addresses the problem of unsupervised detection of recurrent audio segments in TV programs toward program structuring. Recurrent segments are the key elements in the process of program structuring. This allows a direct and non linear access to the main parts inside a program. It hence facilitates browsing within a recorded TV program or a program that is available on a TV-on-Demand service. Our work focuses on programs like entertainments, shows, magazines, news... It proposes an audio recurrence detection method. This is either applied over an episode or a set of episodes of the same program. Different types of audio descriptors are proposed and evaluated over an 65-hours video dataset corresponding to 112 episodes of TV programs.

No abstract

This paper addresses the problem of unsupervised TV program structuring. Program structuring allows direct and non linear access to the desired parts of a program. Our work addresses the structuring of programs like news, entertainment, shows, magazines... It is based on the detection of audio and visual recurrences. It proposes an effective classification and selection system, based on decision trees, that allows the detection of "separators" among these recurrences. Separators are short audio/visual sequences that delimit the different parts of a program. The decision trees are built based on attributes issued from techniques like applause detection, scenes segmentation, face/speaker detection and clustering. The approach has been evaluated on a 112 hours dataset corresponding to 169 episodes of TV programs.

No abstract