Pel-Recursive Motion Estimation Using the Expectation-Maximization Technique and Spatial Adaptation

Vania Estrela⁽¹⁾, Luis A. Rivera⁽¹⁾, Marcos H. S. Bassani⁽²⁾

⁽¹⁾Universidade Estadual do Norte Fluminense, LCMAT-CCT, Grupo de Computação Científica

Av. Alberto Lamego 2000, Campos dos Goytacazes,
RJ, Brasil, CEP 28013-600

{vaniave,rivera}@uenf.br

⁽²⁾Departamento de Engenharia Mecânica, DEPES,CEFET-RJ
Av. Maracanã 229, Rio de Janeiro, RJ, BRAZIL, CEP 20271-110

marcos@cefet-rj.br

ABSTRACT

Pel-recursive motion estimation is a well-established approach. However, in the presence of noise, it becomes an ill-posed problem that requires regularization. In this paper, motion vectors are estimated in an iterative fashion by means of the Expectation-Maximization (EM) algorithm and a Gaussian data model. Our proposed algorithm also utilizes the local image properties of the scene to improve the motion vector estimates following a spatially adaptive approach. Numerical experiments are presented that demonstrate the merits of our methods.

Keywords

Motion estimation, EM algorithm, pel-recursive, maximum likelihood.