Motion Estimation for H.264/AVC using Programmable Graphics Hardware
Department of Computer Science, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China
Multimedia and Expo, 2006 IEEE International Conference on In Multimedia and Expo, 2006 IEEE International Conference on (2006), pp. 2049-2052
@conference{ho2006motion,
title={Motion estimation for H. 264/AVC using programmable graphics hardware},
author={Ho, C.W. and Au, O. and Chan, S.H. and Yip, S.K. and Wong, H.M.},
booktitle={2006 IEEE International Conference on Multimedia and Expo},
pages={2049–2052},
year={2006},
organization={IEEE}
}
We present an efficient implementation of motion estimation (ME) for H.264/AVC using programmable graphics hardware. The cost function for ME in H.264/AVC depends on the motion vector (MV) predictor which is the median MV of three neighboring coded blocks. Previous implementations assume no dependency among adjacent blocks, which is not true for H.264/AVC, they also perform unsatisfactorily because of their low arithmetic intensity, which is defined as operation per word transferred. To overcome the dependency problem, we introduce a new implementation which performs ME on block-by-block basis. Moreover, we can adjust the arithmetic intensity easily to optimize the performance on different graphics cards. Experimental results show that our implementation is substantially faster (by 10 times) than our SIMD optimized CPU implementation.
December 2, 2010 by hgpu
