Video coding using fast geometry-adaptive partitioning and an elastic motion model

  • Authors:

  • Abdullah A. Muhit;Mark R. Pickering;Michael R. Frater;John F. Arnold

  • Affiliations:

  • School of Engineering and Information Technology, University of New South Wales, Australian Defence Force Academy, Canberra, Australia;School of Engineering and Information Technology, University of New South Wales, Australian Defence Force Academy, Canberra, Australia;School of Engineering and Information Technology, University of New South Wales, Australian Defence Force Academy, Canberra, Australia;School of Engineering and Information Technology, University of New South Wales, Australian Defence Force Academy, Canberra, Australia

  • Venue:
  • Journal of Visual Communication and Image Representation
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Effective motion-compensated prediction is the key to high-performance video coding. To ensure continuous improvement of video coders, emerging motion-compensation technologies will need to be successfully integrated into future standards. Higher order elastic motion models and geometry-adaptive block partitioning are such advanced techniques that are good candidates for integration into future generations of video coders. However, it is vital that these techniques are additive in performance, non-interfering and maintain justifiable complexity. In this paper, we propose an efficient block-partitioning scheme that incorporates both geometry-adaptive partitioning and an elastic motion model as extensions to the standard motion estimation procedure. Our experiments suggest that geometric partitioning in combination with the use of an elastic motion model can provide enhanced performance, although the increased complexity is of some concern for real-time applications.