Key frames extraction in athletic video

In this paper, we present an effective framework for features extraction from an athletic sport sequence. We analyze both forward and backward motion vectors from MPEG 2 video sequences for camera movements detection. Features like the beginning and the end of the race and the type of competition are strictly connected to the camera motion. Our algorithm is able to extract the frame number of the investigated feature with very high accuracy.

[1]  Yucel Altunbasak,et al.  Key frame selection from MPEG video data , 1997, Electronic Imaging.

[2]  Jean-Marc Odobez.  Estimation, detection et segmentation du mouvement : une approche robuste et markovienne , 1994 .

[3]  Sanjeev R. Kulkarni,et al.  Rapid estimation of camera motion from compressed video with application to video annotation , 2000, IEEE Trans. Circuits Syst. Video Technol..

[4]  Benoit Mory,et al.  Video motion representation for improved content access , 2000, 2000 Digest of Technical Papers. International Conference on Consumer Electronics. Nineteenth in the Series (Cat. No.00CH37102).

[5]  Sanjeev R. Kulkarni,et al.  A new method for camera motion parameter estimation , 1995, Proceedings., International Conference on Image Processing.

[6]  Edoardo Ardizzone,et al.  Video indexing using optical flow field , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[7]  Yousry S. El Gamal,et al.  Compressed video indexing based on object motion , 2000, Visual Communications and Image Processing.

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

[9]  Boon-Lock Yeo,et al.  Analysis And Presentation Of Soccer Highlights From Digital Video , 1995 .