Proceedings of the 2016 International Conference on Computer Engineering and Information Systems

The Improved Full Search Algorithm for Motion Estimation with GPU Accelection

Authors
Zhuo Liu, Yi-Nan Lu, Tao Liu, Tian-Wen Yuan
Corresponding Author
Zhuo Liu
Available Online November 2016.
DOI
https://doi.org/10.2991/ceis-16.2016.47How to use a DOI?
Keywords
motion estimation; block matching; full search; spatial and temporal correlation; parallel computing
Abstract
Robust motion vector estimation algorithm plays an important role in video processing. Block matching algorithm for motion estimation as a mature search technique has been widely used in video applications because of its simplicity and easy implement. To increase the accuracy of motion vector prediction, an improved motion estimation algorithm is proposed that combines full search algorithm with spatial and temporal correlation. To further reduce the computational cost, a powerful parallel algorithm based on GPU is designed by using a series of optimization methods. Experimental results show that the proposed method gives the high PSNRs by testing the video sequences and the computation speed of the optimized algorithm is greatly improved.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Proceedings
2016 International Conference on Computer Engineering and Information Systems
Part of series
Advances in Computer Science Research
Publication Date
November 2016
ISBN
978-94-6252-283-1
DOI
https://doi.org/10.2991/ceis-16.2016.47How to use a DOI?
Open Access
This is an open access article distributed under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Zhuo Liu
AU  - Yi-Nan Lu
AU  - Tao Liu
AU  - Tian-Wen Yuan
PY  - 2016/11
DA  - 2016/11
TI  - The Improved Full Search Algorithm for Motion Estimation with GPU Accelection
BT  - 2016 International Conference on Computer Engineering and Information Systems
PB  - Atlantis Press
UR  - https://doi.org/10.2991/ceis-16.2016.47
DO  - https://doi.org/10.2991/ceis-16.2016.47
ID  - Liu2016/11
ER  -