3-D motion estimation using a sequence of noisy stereo images

The authors discuss a kinematic-model-based approach for the estimation of 3-D motion parameters from a sequence of noisy stereo images. The approach consists of representing the constant acceleration translational motion and constant angular velocity or constant precession rotational motion in the form of a bilinear state-space model using standard rectilinear states for translation and quaternions for rotation. Closed-form solutions of state transition equations are obtained to propagate the quaternions in both constant-angular-velocity and constant-precession models. The measurements are noisy perturbations of 3-D feature points represented in an inertial coordinate system. It is assumed that the structure is known and that 3-D feature points are extracted from the stereo images and matched over the frames. Owing to the nonlinearity in the state model, nonlinear filters are designed for the estimation of motion parameters