Robust object detection using cascade filter in MPEG videos

We propose a novel approach for motion vector (MV) based object detection in MPEG-2 video streams. Rather than processing the extracted MV fields that are directly extracted from MPEG-2 video streams in the compressed domain, we perform MV smoothing, perform MV noise reduction, obtain more robust object information, and refine this information through a cascaded filter composed of a Gaussian filter and a median filter. As a result, the object detection algorithm is more capable of accurately detecting objects. We compare the performance of our proposed system with the popular and commonly used spatial filter processing techniques: median filter, mean filter, Gaussian filter, and no filter. Based on experimental results performed over the MPEG7 testing dataset and measuring performance using the standard recall and precision metrics, object detection using the cascade filter is remarkably superior to the alternative filtering techniques. In addition to these results, we describe a user system interface that we developed, where users can maintain the filter parameters interactively.

[1]  M. Ibrahim Sezan,et al.  A semantic event-detection approach and its application to detecting hunts in wildlife vide , 2000, IEEE Trans. Circuits Syst. Video Technol..

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

[3]  K. Wakimoto,et al.  Efficient and Effective Querying by Image Content , 1994 .

[4]  Noel E. O'Connor,et al.  Object detection and tracking using an EM-based motion estimation and segmentation framework , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[5]  Kai-Kuang Ma,et al.  Bidirectional motion tracking for video indexing , 1999, 1999 IEEE Third Workshop on Multimedia Signal Processing (Cat. No.99TH8451).

[6]  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..

[7]  N. W. Kim,et al.  Motion analysis using the normalization of motion vectors on MPEG compressed domain , 2002 .

[8]  Roman Goldenberg,et al.  Fast Geodesic Active Contours , 1999, Scale-Space.

[9]  Suh-Yin Lee,et al.  Motion Activity Based Shot Identification and Closed Caption Detection for Video Structuring , 2002, VISUAL.

[10]  HongJiang Zhang,et al.  A new perceived motion based shot content representation , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[11]  Gunnar Farnebäck.  Motion-based segmentation of image sequences , 1996 .

[12]  Hua-Tsung Chen,et al.  Motion Activity Based Semantic Video Similarity Retrieval , 2002, IEEE Pacific Rim Conference on Multimedia.

[13]  Richard Szeliski,et al.  An integrated Bayesian approach to layer extraction from image sequences , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[14]  David S. Doermann,et al.  Building mosaics from video using MPEG motion vectors , 1999, MULTIMEDIA '99.

[15]  Liang-Gee Chen,et al.  Efficient moving object segmentation algorithm using background registration technique , 2002, IEEE Trans. Circuits Syst. Video Technol..

[16]  Suh-Yin Lee,et al.  Motion-Based Semantic Event Detection for Video Content Description in MPEG-7 , 2001, IEEE Pacific Rim Conference on Multimedia.

[17]  Ya-Qin Zhang,et al.  A confidence measure based moving object extraction system built for compressed domain , 2000, 2000 IEEE International Symposium on Circuits and Systems. Emerging Technologies for the 21st Century. Proceedings (IEEE Cat No.00CH36353).

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

[19]  Andrew Blake,et al.  Dynamic contours: real-time active splines , 1993 .

[20]  R. Ulichney.  37.4: Filter Design for Void-and-Cluster Dither Arrays , 1997 .

[21]  Rangasami L. Kashyap,et al.  Video scene change detection method using unsupervised segmentation and object tracking , 2001, IEEE International Conference on Multimedia and Expo, 2001. ICME 2001..

[22]  R. Venkatesh Babu,et al.  Compressed domain motion segmentation for video object extraction , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[23]  D. P. Elias,et al.  The motion-based segmentation of image sequences , 1999 .

[24]  Nuno Vasconcelos,et al.  Empirical Bayesian EM-based motion segmentation , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[25]  Christos Faloutsos,et al.  Efficient and effective Querying by Image Content , 1994, Journal of Intelligent Information Systems.

[26]  Paul W. Fieguth,et al.  Color-based tracking of heads and other mobile objects at video frame rates , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[27]  Javed I. Khan,et al.  Motion based object tracking in MPEG-2 stream for perceptual region discriminating rate transcoding , 2001, MULTIMEDIA '01.

[28]  Hiroshi Murase,et al.  Video shot analysis using efficient multiple object tracking , 1997, Proceedings of IEEE International Conference on Multimedia Computing and Systems.

[29]  Shih-Fu Chang,et al.  Compressed-domain techniques for image/video indexing and manipulation , 1995, Proceedings., International Conference on Image Processing.

[30]  Gerald Kühne,et al.  Motion-based segmentation and contour-based classification of video objects , 2001, MULTIMEDIA '01.