Adaptive macro motion vector quantization

  • Authors:

  • Luis A. da Silva Cruz

  • Affiliations:

  • Dep. of Electrical and Computer Engineering, Polo 2 University of Coimbra, Coimbra, Portugal

  • Venue:
  • PCM'04 Proceedings of the 5th Pacific Rim Conference on Advances in Multimedia Information Processing - Volume Part II
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Block based motion estimation can be interpreted as a form of vector quantization (VQ) based on a codebook of motion vectors. As is usually the case in VQ the codebook can be precomputed using the Loyd-Buzzo-Gray algorithm and a set of training data or generated on-the-fly in an adaptive fashion. In this work we present an adaptive motion estimation method based on vector quantization of blocks of motion vectors with a dynamically updated codebook which reduces the complexity of the motion estimation process and the motion side information.