N. Tagawa et al., ESTIMATION OF 3-D MOTION FROM OPTICAL-FLOW WITH UNBIASED OBJECTIVE FUNCTION, IEICE transactions on information and systems, E77D(10), 1994, pp. 1148-1161
This paper describes a noise resistant algorithm for estimating 3-D ri
gid motion from optical flow. We first discuss the problem of construc
ting the objective function to be minimized. If a Gaussian distributio
n is assumed for the noise, it is well-known that the least-squares mi
nimization becomes the maximum likelihood estimation. However, the use
of this objective function makes the minimization procedure more expe
nsive because the program has to go through all the points in the imag
e at each iteration. We therefore introduce an objective function that
provides unbiased estimators. Using this function reduces computation
al costs. Furthermore, since good approximations can be analytically o
btained for the function, using them as an initial guess we can apply
an iterative minimization method to the function, which is expected to
be stable. The effectiveness of this method is demonstrated by comput
er simulation.