A novel formulation of the visual motion estimation problem is analyzed in terms of identification of the nonlinear implicit system with parameters on the so called “essential manifold”; and a new and unified implicit adaptive extended kalman filter (IAEKF) is proposed to perform the estimation in the local coordinates, for both discrete and continuous cases. The correctness of the algorithm is illustrated by numerical simulations.