3-D INTERPRETATION OF OPTICAL-FLOW BY RENORMALIZATION

This article studies 3-D interpretation of optical flow induced by a general camera motion relative to a surface of general shape. First, we describe, using the ''image sphere representation,'' an analytical procedure that yields an exact solution when the data are exact: we solve the epipolar equation written in terms of the essential parameters and the twisted optical flow. Introducing a simple model of noise, we then show that the solution is ''statistically biased.'' In order to remove the statistical bias, we propose an algorithm called renormalization, which automatically adjusts to unknown image noise. A brief discussion is also given to the critical surface that yields ambiguous 3-D interpretations and the use of the image plane representation.