The motion recovery for a class of movements in the space is considered in this note by using the stereo perspective observation of at least three points. The motion equation that is considered can cover a wide class of practical movements in the space. The asymptotic estimation of the motion parameters, which are all time-dependent variables, is developed based on the second method of Lyapunov. Furthermore, the proposed algorithm can be modified to deal with occlusion phenomenon. To illustrate the performance of the proposed algorithm, it is compared with the extended Kalman filter (EKF) by computer simulations.