An Estimation-Theoretic Framework for Image-Flow Computation

Abstract

Image-flow is a major source of three-dimensional information. This paper describes a new framework for computing image-flow from time-varying imagery. In this framework, image-flow information is classified into two categories - conservation information and neighborhood information. Each type of information is recovered in the form of an estimate accompanied by a covariance-matrix. Image-flow is then computed by fusing the two estimates using estimation-theoretic techniques. This framework offers the following principal advantages. Firstly, it allows estimation of certain types of discontinuous flow-fields without any a-priori knowledge about the location of discontinuities. The flow-fields thus recovered are not blurred at motion-boundaries. Secondly, covariance matrices (or alternatively. confidence-measures) are associated with the estimate of image-flow at, each stage of computation. The estimation-theoretic nature of the framework and its ability to provide covariance matrices make it very useful in the context of applications such as incremental estimation of scene-depth using techniques based on Kalman filtering. In this paper, an algorithm based on this framework is used to recover image-flow from two image-sequences. To illustrate an application, the image-flow estimates and their covariance matrices thus obtained are also used to recover scene-depth.