An Efficient Parallelization Scheme of Motion Estimation Algorithms on Myrinet-connected Workstations

 Motion algorithms have been developed for various applications such as image sequence analysis, machine vision, robotics, motion picture restoration or video coding. Motion estimation requires high processing requirements to estimate the image sequence with a reasonable frame size and quality. In this paper, we present an efficient parallelization scheme of motion estimation algorithms on Myrinet-connected workstations. Results indicated that using the proposed parallelization scheme, almost linear speedup curve could be achieved for four HDTV test video sequences experimented, indicating that if more processors are involved, less time will be required to process a frame.

[1]  Fabio Bellifemine,et al.  Evolution and trends of HDTV , 1991, [1991] Proceedings, Advanced Computer Technology, Reliable Systems and Applications.

[2]  A. Kokaram.  Motion picture restoration , 1998 .

[3]  Teddy Surya Gunawan,et al.  Performance of the Communication Layer with the Myrinet Gigabit LAN for HDTV Images on Pentium-Linux cluster using MPI , 2000 .

[4]  Konstantinos Konstantinides,et al.  Image and video compression standards , 1995 .

[5]  Amnon Barak,et al.  The MOSIX Distributed Operating System: Load Balancing for UNIX , 1993 .

[6]  Charles L. Seitz,et al.  Myrinet: A Gigabit-per-Second Local Area Network , 1995, IEEE Micro.

[7]  C. L. Lee.  Parallel implementation of motion-compensation for HDTV video decoder , 1997, ISCE '97. Proceedings of 1997 IEEE International Symposium on Consumer Electronics (Cat. No.97TH8348).

[8]  Ishfaq Ahmad,et al.  Performance of software-based MPEG-2 video encoder on parallel and distributed systems , 1997, IEEE Trans. Circuits Syst. Video Technol..

[9]  Teddy Surya Gunawan,et al.  Motion estimation using adaptive blocksize observation model and efficient multiscale regularization , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).

[10]  M. Bierling,et al.  Displacement Estimation By Hierarchical Blockmatching , 1988, Other Conferences.

[11]  Martin Vetterli,et al.  A multiresolution approach to motion estimation and interpolation with application to coding of digital HDTV , 1990, IEEE International Symposium on Circuits and Systems.

[12]  Jongho Nang,et al.  An effective parallelizing scheme of MPEG-1 video encoding on Ethernet-connected workstations , 1997, Proceedings. Advances in Parallel and Distributed Computing.

[13]  Ioannis Pitas,et al.  Parallel Algorithms for Digital Image Processing, Computer Vision and Neural Networks , 1993 .

[14]  Frederic Dufaux,et al.  Motion estimation techniques for digital TV: a review and a new contribution , 1995, Proc. IEEE.

[15]  Man-Nang Chong,et al.  Bidirectional motion estimation via vector propagation , 1998, IEEE Trans. Circuits Syst. Video Technol..

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

[17]  Dilip Krishnan,et al.  Video Restoration on a Multiple TMS320C40 System , 1996 .

[18]  Teddy Surya Gunawan,et al.  Motion estimation using adaptive matching and multiscale methods , 2000, Visual Communications and Image Processing.

[19]  Janak H. Patel,et al.  Parallel implementation and evaluation of motion estimation system algorithms on a distributed memory multiprocessor using knowledge based mappings , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.