Open Access Open Access  Restricted Access Subscription or Fee Access

Implementation and Optimization of Full Search Motion Estimation Algorithm on CUDA

F. E. Sayadi, M. Chouchene, H. Bahri, R. Khemiri, M. Atri

Abstract



Demand for high quality video is growing exponentially especially with the H.264/AVC and the H.265/HEVC standards which have placed a significant emphasis on computational power of the processors. In both standards, the motion estimation block takes very intensive processing. That is why recently Graphics Processing Units have emerged as a viable target for speeding such applications by harnessing parallelism.
This paper describes our GPU Full Search Motion Estimation implementation and presents novel methods of optimization, allowing to manipulate the GPU in a flexible way. These optimization methods can provide high performance video encoder where the speed up reaches about 90.

Keywords


GPU, CUDA, Nvidia optimization method, Motion Estimation: Full Search algorithm.

Full Text:

PDF


Disclaimer/Regarding indexing issue:

We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.