このエントリーをはてなブックマークに追加
ID 44496
フルテキストURL
著者
Kanatani, Kenichi Department of Computer Science, Okayama University
Rangrajan, Prasanna Department of Electrical Engineering, Southern Methodist University
Sugaya, Yasuyuki Department of Computer Science and Engineering, Toyohashi University of Technology
Niitsuma, Hirotaka Department of Computer Science, Okayama University
抄録
We present a new least squares (LS) estimator, called “HyperLS”, specifically designed for parameter estimation in computer vision applications. It minimizes the algebraic distance under a special scale normalization, which is derived by rigorous error analysis in such a way that statistical bias is removed up to second order noise terms. Numerical experiments suggest that our HyperLS is far superior to the standard LS and comparable in accuracy to maximum likelihood (ML), which is known to produce highly accurate results in image applications but may fail to converge if poorly initialized. Our HyperLS is a perfect candidate for ML initialization. In addition, we discuss how image-based inference problems have different characteristics form conventional statistical applications, with a view to serving as a bridge between mathematicians and computer engineers.
発行日
2011-01
出版物タイトル
Memoirs of the Faculty of Engineering, Okayama University
出版物タイトル(別表記)
岡山大学工学部紀要
45巻
出版者
Faculty of Engineering, Okayama University
開始ページ
15
終了ページ
26
ISSN
1349-6115
NCID
AA12014085
資料タイプ
紀要論文
言語
English
著作権者
Copyright © by the authors
論文のバージョン
publisher
査読
無し
Eprints Journal Name
mfe