Premium
Residual‐Based robust estimation and image‐motion analysis
Author(s) -
Zhuang Xinhua,
Zhao Yunxin,
Huang Thomas S.
Publication year - 1990
Publication title -
international journal of imaging systems and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.359
H-Index - 47
eISSN - 1098-1098
pISSN - 0899-9457
DOI - 10.1002/ima.1850020411
Subject(s) - outlier , estimator , artificial intelligence , computer science , robustness (evolution) , motion estimation , residual , heuristics , computer vision , segmentation , robust statistics , feature (linguistics) , range (aeronautics) , mathematics , pattern recognition (psychology) , algorithm , statistics , biochemistry , chemistry , linguistics , philosophy , materials science , composite material , gene , operating system
We explain that the task of multiple rigid motion segmentation and estimation from image feature point correspondence demands an estimator of high robustness. We show that a heuristics‐based partial modeling approach can be used to develop a highly robust estimator called the MF estimator for general regression, where “MF” represents an abbreviation of Model Fitting. Finally, we provide experimental results in estimating single rigid motion from a mixture of 2D‐2D (or image‐image), 3D‐2D (or range‐image) and 3D‐3D (or range‐range) corresponding point data by using the proposed MF estimator. As will be seen, only four well matched corresponding point pairs are needed to get a good estimate of motion parameters no matter how many mismatched corresponding point pairs (or outliers) occur. The article represents an initial effort towards robust image‐motion analysis.