Video summarization with moving objects in the compressed domain

The vast amount of video sequences avaliable in digital format presents considerable challenges for descriptor extraction and information retrieval. The dominant motion in a video scene proves to be very important to characterize video sequences, but the cost to compute it is high when working in image domain. In this paper we present a method to extract an affine description of the global motion of a video sequence using a robust estimator and data from the compressed domain, where the motion vector field is already calculated. We also isolate and describe parametrically the local motions using non parametric clustering. Finally, we apply our approach to motion-based segmentation of video sequences and video summarization.

[1]  Dorin Comaniciu,et al.  Robust analysis of feature spaces: color image segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[2]  Michal Irani,et al.  Video indexing based on mosaic representations , 1998, Proc. IEEE.

[3]  C. D. Kemp,et al.  Density Estimation for Statistics and Data Analysis , 1987 .

[4]  P. Rousseeuw.  Least Median of Squares Regression , 1984 .

[5]  Peter Meer,et al.  ROBUST TECHNIQUES FOR COMPUTER VISION , 2004 .

[6]  Yanjun Qi,et al.  A probabilistic model for camera zoom detection , 2002, Object recognition supported by user interaction for service robots.

[7]  Shih-Fu Chang,et al.  Survey of compressed-domain features used in audio-visual indexing and analysis , 2003, J. Vis. Commun. Image Represent..

[8]  Sánchez Secades,et al.  Multiple feature temporal models for the characterization of semantic video contents , 2003 .

[9]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  Dorin Comaniciu,et al.  Real-time tracking of non-rigid objects using mean shift , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[11]  Alain Crouzil,et al.  Non-rigid object localization from color model using mean shift , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[12]  Ilan Shimshoni,et al.  Mean shift based clustering in high dimensions: a texture classification example , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[13]  Bo Yu.  Recognition of freehand sketches using mean shift , 2003, IUI '03.

[14]  Larry D. Hostetler,et al.  The estimation of the gradient of a density function, with applications in pattern recognition , 1975, IEEE Trans. Inf. Theory.

[15]  Thomas S. Huang,et al.  Fast camera motion analysis in MPEG domain , 1999, Proceedings 1999 International Conference on Image Processing (Cat. 99CH36348).

[16]  Yizong Cheng,et al.  Mean Shift, Mode Seeking, and Clustering , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Patrick Gros,et al.  A Geometrical Key-Frame Selection Method Exploiting Dominant Motion Estimation in Video , 2004, CIVR.

[18]  Christos Faloutsos,et al.  Compressed-domain video indexing techniques using DCT and motion vector information in MPEG video , 1997, Electronic Imaging.

[19]  David Suter,et al.  Variable bandwidth QMDPE and its application in robust optical flow estimation , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[20]  D. Comaniciu,et al.  The variable bandwidth mean shift and data-driven scale selection , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[21]  C. Stiller,et al.  Estimating Motion in Image Sequences A tutorial on modeling and computation of 2 D motion , 2022 .