COMPRESSED-DOMAIN OBJECT DETECTION FOR VIDEO UNDERSTANDING

In this paper, a novel algorithm for the real-time, unsupervised object detection in compressed-domain sequences is proposed. The algorithm utilizes color and motion information present in the compressed stream as well as a simple object model. Extraction of the MPEG-7 dominant color descriptor, clustering of macroblocks to dominant color clusters and model-based cluster selection are employed for object detection in I-frames, while temporal tracking is employed for P-frames. The proposed methodology assumes neither a static camera nor that there exists a single dominant color in the frame, which represents the object of interest. Experimental results of road detection in various sequences demonstrate the efficiency of the proposed approach and reveal the potential of employing it for video understanding and semantic information extraction in contextspecific applications.

[1]  Masahito Hirakawa,et al.  Knowledge-assisted content-based retrieval for multimedia databases , 1994, 1994 Proceedings of IEEE International Conference on Multimedia Computing and Systems.

[2]  Leonidas J. Guibas,et al.  A metric for distributions with applications to image databases , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[3]  Lorenzo Favalli,et al.  Object tracking for retrieval applications in MPEG-2 , 2000, IEEE Trans. Circuits Syst. Video Technol..

[4]  B. S. Manjunath,et al.  Color and texture descriptors , 2001, IEEE Trans. Circuits Syst. Video Technol..

[5]  Thomas Sikora,et al.  The MPEG-7 visual standard for content description-an overview , 2001, IEEE Trans. Circuits Syst. Video Technol..

[6]  Michael G. Strintzis,et al.  A framework for the efficient segmentation of large-format color images , 2002, Proceedings. International Conference on Image Processing.

[7]  Shih-Fu Chang,et al.  The holy grail of content-based media analysis , 2002 .

[8]  Michael G. Strintzis,et al.  An ontology approach to object-based image retrieval , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[9]  A. Murat Tekalp,et al.  Automatic Soccer Video Analysis and Summarization , 2003, IS&T/SPIE Electronic Imaging.

[10]  Michael G. Strintzis,et al.  REAL-TIME COMPRESSED-DOMAIN SPATIOTEMPORAL VIDEO SEGMENTATION , 2003 .

[11]  Michael G. Strintzis,et al.  Video object segmentation using Bayes-based temporal tracking and trajectory-based region merging , 2004, IEEE Transactions on Circuits and Systems for Video Technology.