Optimal projection of 2-D displacements for 3-D translational motion estimation

被引：6

作者：

Garcia, C ^{[1
]}

Tziritas, G ^{[1
]}

机构：

[1] Univ Crete, Dept Comp Sci, Iraklion, Greece

来源：

IMAGE AND VISION COMPUTING | 2002年 / 20卷 / 11期

关键词：

motion estimation; optical flow; unbiased estimator;

D O I：

10.1016/S0262-8856(02)00088-4

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Recovering 3-D motion parameters from 2-D displacements is a difficult task, given the influence of noise contained in these data, which correspond at best to a crude approximation of the real motion field. Stability for the system of equations to solve is therefore essential. In this paper, we present a novel method based on an unbiased estimator that aims at enhancing this stability and strongly reduces the influence of noise contamination. Experimental results using synthetic and real optical flows are presented to demonstrate the effectiveness of our method in comparison to a set of selected methods. (C) 2002 Published by Elsevier Science B.V.

引用

页码：793 / 804

页数：12

共 19 条

[1] DETERMINING 3-DIMENSIONAL MOTION AND STRUCTURE FROM OPTICAL-FLOW GENERATED BY SEVERAL MOVING-OBJECTS [J].