Recursive estimation of motion and shape parameters under perspective projection

Abstract

The problem of estimating motion and shape of a moving rigid object has been considered from a time series of image data obtained via perspective projection. First it is shown that, for a class of motion parameters, the shape parameters are asymptotically decoupled from the optical flow dynamics, hence become unobservable asymptotically. For other choices of motion parameters, shape parameters remain asymptotically obseravble. Secondly we show that, if the surface is almost planar, then uncertainty in the shape of the surface propagates as an optical flow equation with additive state dependent uncertainty. Finally, we develop a recursive estimation algorithm, and present the numerical simulation results.