본문 바로가기
  • Home

An Enhanced Search Algorithm for Fast Motion Estimation using Sub-Pixel

  • Journal of The Korea Society of Computer and Information
  • Abbr : JKSCI
  • 2011, 16(12), pp.103-112
  • Publisher : The Korean Society Of Computer And Information
  • Research Area : Engineering > Computer Science

김대곤 1 Yoo Cheol Jung ORD ID 1

1전북대학교

Accredited

ABSTRACT

Motion estimation (ME) is regarded as an important component in a video encoding process, because it consumes a large computation complexity. H.264/AVC requires additional computation overheads for fractional search and interpolation. This causes a problem that computational complexity is increased. In Motion estimation, SATD(Sum of Transform Difference) has the characteristics of a parabolic based on the minimum point. In this paper, we propose new prediction algorithm to reduce search point in motion estimation by sub-pixel interpolation characteristics. The proposed algorithm reduces the time of encoding process by decreasing computational complexity. Experimental results show that the proposed method reduces 20% of the computation complexity of motion estimation, while the degradation in video quality is negligible.

Citation status

* References for papers published after 2023 are currently being built.