Hierarchical model based robust global motion estimation technique

Global motion caused by camera vibration is quite common in image sequence. It can be modeled by a few parameters and can be estimated using regression technique which first estimates the local motion and then uses the local information to find the global motion that minimize the least square error. Then the global motion is compensated using these parameters. Refinements are carried out during the estimation inner process based on iterative elimination of singular values which introduces the bias to the estimated parameters. In order to compensate efficiently the larger image motions, the hierarchical process is introduced. The basic idea of hierarchical processing is to form a pyramid of image sequences so that on each pyramid level the motion is small enough to be estimated and the motion on the original data can be calculated as a weighted sum of results on all levels. Theoretical analysis of the technique's performance and experimental results are also given.