Recovering of visual scenarios in movies by motion analysis and grouping spatio-temporal colour signatures of video shots

The paper proposes a new, ID signal approach to basic operations required for video structuring and video content description: motion estimation, segmenting into shots, and scene grouping. The approach is based on the projective transform called Mojette which is a discrete analog of Radon transform.

[1]  R. Gray,et al.  Vector quantization , 1984, IEEE ASSP Magazine.

[2]  Riccardo Leonardi,et al.  Describing multimedia documents in natural and semantic-driven ordered hierarchies , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[3]  Alberto Del Bimbo,et al.  Visual information retrieval , 1999 .

[4]  Boon-Lock Yeo,et al.  Time-constrained clustering for segmentation of video into story units , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[5]  Jenny Benois-Pineau,et al.  Image Segmentation in compressed domain by clustering methods with euclidean and p-adic metrics , 2002 .

[6]  Jenny Benois-Pineau,et al.  Dominant motion estimation and video partitioning with a 1D signal approach , 1998, Other Conferences.

[7]  Philippe Joly,et al.  Efficient automatic analysis of camera work and microsegmentation of video using spatiotemporal images , 1996, Signal Process. Image Commun..

[8]  Patrick Bouthemy,et al.  A unified approach to shot change detection and camera motion characterization , 1999, IEEE Trans. Circuits Syst. Video Technol..