A contour-based recovery of image flow: iterative method

The authors present an iterative algorithm for the recovery of 2D motion, i.e. an algorithm for the determination of a transformation which maps one image onto another. The local ambiguity in measuring the motion of contour segments (called the aperture problem) makes it necessary to rely on measurements along the normal direction. Since the measured normal flow itself does not agree with the actual normal flow, the full flow recovered from this erroneous normal flow possesses substantial error too, and any attempt to recover the 3D motion from such full flow is doomed to failure. The present method is based on the observation that a polynomial approximation of image flow provides sufficient information for 3D motion computation. The use of an explicit flow model makes it possible to improve normal flow estimates through an iterative process. The algorithm has been tested on synthetic and simple natural time-varying images. The image flow recovered from this scheme was sufficiently accurate as to be useful in 3D structure and motion computation