Motion vector extrapolation for parallel motion estimation on GPU

Yi Gao

doi:10.1007/s11042-012-1074-4

Return

Motion vector extrapolation for parallel motion estimation on GPU

2012-03-31

PRE

Yi Gao *

Jun Zhou

DOI：10.1007/s11042-012-1074-4

Original

Original request for help

Save

Abstract

En 中文

The powerful parallel computing ability of Graphics Processing Unit (GPU) has shown its striking superiority for motion estimation acceleration in conventional hybrid video encoding process. Unfortunately, the motion information of the neighboring macroblocks is not available for current macroblock, such that parallel motion estimation using GPU is not very favored. To tackle this problem while achieving high acceleration ration, motion vector cost is always ignored in most existing solutions, which inevitably causes severe rate-distortion loss. In this paper, a novel motion vector extrapolation based approach (MVEA) is presented for enhancing rate-distortion performance of parallel motion estimation on GPU, which is based on the study of motion vector recovery strategies for frame loss error concealment. Furthermore, the efficient implementation of MVEA on Computing Unified Device Architecture (CUDA) is also investigated. Simulation results show that MVEA can achieve a maximum peak Signal-to-Noise ratio enhancement of 0.8 dB with ignorable computational cost increase.

Keywords:

Motion estimation

Error concealment

GPU

CUDA

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.