Spline-Based Image Registration

  • Authors:
  • Richard Szeliski;James Coughlan

  • Affiliations:
  • Microsoft Research, One Microsoft Way, Redmond, WA 98052-6399/ E-mail: szeliski@microsoft.com;Harvard University, Department of Physics, Cambridge, MA 02138/ E-mail: coughlan@hrl.harvard.edu

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

The problem of image registration subsumes a number of problems andtechniques in multiframe image analysis, including the computation of opticflow (general pixel-based motion), stereo correspondence,structure from motion, and feature tracking.We present a new registration algorithm based on spline representationsof the displacement field which can be specialized to solveall of the above mentioned problems. In particular, we show how to computelocal flow, global (parametric) flow, rigid flow resulting from cameraegomotion, and multiframe versions of the above problems. Using a spline-based description of the flow removes the need foroverlapping correlation windows, and produces an explicit measure of thecorrelation between adjacent flow estimates.We demonstrate our algorithm on multiframe image registrationand the recovery of 3D projective scene geometry.We also provide results on a number of standard motion sequences.