Optic Flow Computation
Sequences of time ordered images allow the estimation of projected two-dimensional image motion as either instantaneous image velocities or discrete image displacements. This is referred to as the optical flow (or image velocity) field. The computation of optical flow is streamlined when referring to an imaging system. Here, the vector field describes the movement of texture across the camera's image plane and is thus a two-dimensional problem.
The underlying assumption that is used to compute optical flow is that the brightness (or image intensity, I) of objects in the environment is constant over time (see Figure 1).
In other words, the derivative of the image intensity, dI/dt, is zero:
The approximation of vx and vy is not easily solved - one equation and two unknowns. This is primarily because given the spatioal derivatives, Ix and Iy, and the termporal derivative, It, there is a locus of possible velocities (vx,vy) that could be present. This is known as the motion constraint line.