Title :
Parametric object motion estimation
Author :
Stiller, Christoph ; Suntrup, Ralph
Author_Institution :
Inst. for Commun. Eng., Aachen Univ. of Technol., Germany
Abstract :
The authors are concerned with parametric motion estimation for 2-D objects in image sequences. They propose a generalized parametric motion model covering several of the commonly used motion models as special cases. Two algorithms for parameter estimation are presented. The first algorithm is a gradient technique closely related to the one proposed by Diehl (1991). The second is a region matching evolution technique. Simulation results demonstrate that the generalized parameteric model improves performance of motion compensation as compared with traditional motion models
Keywords :
motion estimation; parameter estimation; 2-D objects; gradient technique; image sequences; motion compensation; parameter estimation algorithms; parametric motion estimation; parametric motion model; region matching evolution; simulation results; Image processing; Image segmentation; Image sequences; Layout; Motion compensation; Motion estimation; Object oriented modeling; Parameter estimation; Parametric statistics; State-space methods;
Conference_Titel :
Singapore ICCS/ISITA '92. 'Communications on the Move'
Print_ISBN :
0-7803-0803-4
DOI :
10.1109/ICCS.1992.255187
