Video content representation using optimal extraction of frames and scenes

An efficient video content representation is proposed using optimal extraction of characteristic frames and scenes. This representation, apart from providing browsing capabilities to digital video databases, also allows more efficient content-based queries and indexing. For performing the frame/scene extraction, a feature vector formulation of the images is proposed based on color and motion segmentation. Then, the scene selection is accomplished by clustering similar scenes based on a distortion criterion. Frame selection is performed using an optimization method for locating a set of minimally correlated feature vectors.

[1]  S. Kollias,et al.  INDEXING AND RETRIEVAL OF THE MOST CHARACTERISTIC FRAMES / SCENES IN VIDEO DATABASES , 1997 .

[2]  Shih-Fu Chang,et al.  Joint adaptive space and frequency basis selection , 1997, Proceedings of International Conference on Image Processing.

[3]  Giridharan Iyengar,et al.  VideoBook: an experiment in characterization of video , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[4]  Ahmed H. Tewfik,et al.  Eigen-image based video segmentation and indexing , 1997, Proceedings of International Conference on Image Processing.

[5]  Bernard Merialdo,et al.  Automatic indexing of TV News , 1997 .

[6]  Michael Mills,et al.  Blockmatching motion estimation algorithms-new results , 1990 .

[7]  Thomas S. Huang,et al.  Content-based image retrieval with relevance feedback in MARS , 1997, Proceedings of International Conference on Image Processing.

[8]  Allen Gersho,et al.  Vector quantization and signal compression , 1991, The Kluwer international series in engineering and computer science.

[9]  Shih-Fu Chang,et al.  Digital image/video library and MPEG-7: standardization and research issues , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).