Abstract
The recovery of the three-dimensional motion parameters and structure of an object from its perspectively projected two-dimensional dynamic picture image is one of the most important problems in computer vision. Previous works on shape and motion from optical flow have been unsatisfactory because they require solving nonlinear simultaneous equations using iterative search. In the present paper, we propose a method for recovering motion and structure from perspectively projected optical flow of feature points which move rigidly. In order to simplify the computation and obtain clear perspectives, we adopt a spherical image screen and arrange equations in linear form. As a result, our method does not require the solution of nonlinear simultaneous equations, but only requires the solution of linear ones. In addition, the uniqueness of the solution is guaranteed. Furthermore, we propose a reliability measure of the recovery. Simulation results of our method are also presented.
