Global motion vector estimation via elaborated Reichardt parameter optimization

  • Authors:

  • Eiji Nakamura;Chikara Takahashi;Katsutoshi Sawada

  • Affiliations:

  • Aichi Institute of Technology, Yakusa Toyota, Aichi, Japan;Aichi Institute of Technology, Yakusa Toyota, Aichi, Japan;Aichi Institute of Technology, Yakusa Toyota, Aichi, Japan

  • Venue:
  • SIP '07 Proceedings of the Ninth IASTED International Conference on Signal and Image Processing
  • Year:
  • 2007

Quantified Score

Hi-index 0.02

Visualization

Abstract

This paper resents a novel global motion vector estimation algorithm based on the so called Elaborated Reichardt Detector or ERD. The algorithm optimizes numerically parameters defining temporal delay and spatial filters for every image frame observed by the ERD. Based on the obtained optimal parameters, the speed and direction of the observed image are estimated. It has been found that velocities can be estimated with great accuracy for different types of image sequences. Also, the computational cost of the algorithm has been found to be very low regardless of the numerical optimization process performed.