A New Approach To Content-Based Video Indexing Using Hidden Markov Models

This paper presents one of the rst investigations of the use of Hidden Markov Models (HMMs) for content-based video indexing. A real-time video indexing system for TV news is described, which is able to distinguish between six diierent classes. The huge potential of these methods for video indexing of movies and TV shows is demonstrated by presenting the rst very promising results.

[1]  Gerhard Rigoll,et al.  New improved feature extraction methods for real-time high performance image sequence recognition , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[3]  Wolfgang Effelsberg,et al.  Automatic recognition of film genres , 1995, MULTIMEDIA '95.

[4]  Gerhard Rigoll,et al.  A new approach to video sequence recognition based on statistical methods , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[5]  Yasuo Ariki,et al.  Extraction of TV news articles based on scene cut detection using DCT clustering , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.