A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Performance and Complexity Analysis of Motion Estimation Using Multiple Constraints in Video Compression
Abstract One of the fast rapidly growing fields during the last thirty years is image and video compression, resulting in the development of various video coding standards. The major objective of developing these standards is to attain low bit rate for data storage and transmission, while maintaining acceptable video quality. Motion estimation which is used for estimating the motion between the current and reference frames is key component of these standards. It takes 80–90% time of a video encoder since it is computationally expensive and has a significant impact on the output video quality. Block based motion estimation algorithms are used for providing optimal output but due to computational complexity, it is rarely used in real time application. In this paper, we have presented optimal edge threshold based quad tree motion estimator decomposition for motion estimation by adjusting Edge Threshold to achieve trade-off between computational complexity and video quality (PSNR) in video compression. To solve the problem of bit allocation, Lagrangian Multiplier update algorithms is used imposing multiple constraints such rate and distortion for finding multi constraint shortest paths. A* prune algorithm provides the best optimal solution to R-D optimization problem of video compression by selecting K-multi constraints shortest path. It is shown by experimental simulation that proposed quad tree algorithms along with A* prune algorithms reduces bit rate to 30–40% of traditional quad tree algorithms with other rate distortion optimization technique.
Performance and Complexity Analysis of Motion Estimation Using Multiple Constraints in Video Compression
Abstract One of the fast rapidly growing fields during the last thirty years is image and video compression, resulting in the development of various video coding standards. The major objective of developing these standards is to attain low bit rate for data storage and transmission, while maintaining acceptable video quality. Motion estimation which is used for estimating the motion between the current and reference frames is key component of these standards. It takes 80–90% time of a video encoder since it is computationally expensive and has a significant impact on the output video quality. Block based motion estimation algorithms are used for providing optimal output but due to computational complexity, it is rarely used in real time application. In this paper, we have presented optimal edge threshold based quad tree motion estimator decomposition for motion estimation by adjusting Edge Threshold to achieve trade-off between computational complexity and video quality (PSNR) in video compression. To solve the problem of bit allocation, Lagrangian Multiplier update algorithms is used imposing multiple constraints such rate and distortion for finding multi constraint shortest paths. A* prune algorithm provides the best optimal solution to R-D optimization problem of video compression by selecting K-multi constraints shortest path. It is shown by experimental simulation that proposed quad tree algorithms along with A* prune algorithms reduces bit rate to 30–40% of traditional quad tree algorithms with other rate distortion optimization technique.
Performance and Complexity Analysis of Motion Estimation Using Multiple Constraints in Video Compression
Kumar, Rajender (author) / Kumar, Krishan (author) / Pandit, Amit Kant (author)
2019-06-28
9 pages
Article/Chapter (Book)
Electronic Resource
English
SOCIO-ORGANO COMPLEXITY, PROJECT SCHEDULE PERFORMANCE AND UNDERDAMPED TRANSIENT MOTION
| BASE | 2020
| TIBKAT | 1999
| British Library Conference Proceedings | 1997