Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Direct estimation of long-term global motion parameters using affine and higher order polynomial models

: Smolic, A.; Sikora, T.; Ohm, J.-R.

Tabatabai, A.:
Picture Coding Symposium '99 : April 21 - 23, 1999, Portland Hilton, Portland, Oregon
Davis: University of California, 1999
Picture Coding Symposium <1999, Portland/Or.>
Fraunhofer HHI ()
feature extraction; image coding; image matching; image sequences; motion estimation; parameter estimation; polynomials; prediction theory; long-term global motion parameters; affine polynomial models; higher order polynomial models; image objects; time-consistent motion; sequence; parabolic model; feature matching; optical flow; hierarchical strategy; translation; closed-loop prediction; mpeg-4; sprite coding

We present a new technique for the long-term global motion estimation of image objects. The estimated motion parameters describe the continuous and time-consistent motion over the whole sequence relative to a fixed reference co-ordinate system. The proposed method is suitable for the estimation of affine motion parameters as well as for higher-order motion models like the parabolic model. It combines the advantages of feature matching and optical flow techniques. A hierarchical strategy is applied for the estimation, first translation, then affine motion and finally higher-order motion parameters, which is robust and computationally efficient. A closed-loop prediction scheme is applied to avoid the problem of error accumulation in long-term motion estimation. The presented results indicate that the proposed technique is a very accurate and robust approach for long-term global motion estimation, which can be used for applications such as MPEG-4 sprite coding. We also show that the efficiency of global motion estimation can be significantly increased if a higher-order motion model is applied.