| Item type |
SIG Technical Reports(1) |
| 公開日 |
2016-02-25 |
| タイトル |
|
|
タイトル |
Computation of Upper and Lower Bounds of <i>L</i><sub>2</sub> Error of Multiview Triangulation Using Linear Matrix Inequalities |
| タイトル |
|
|
言語 |
en |
|
タイトル |
Computation of Upper and Lower Bounds of <i>L</i><sub>2</sub> Error of Multiview Triangulation Using Linear Matrix Inequalities |
| 言語 |
|
|
言語 |
eng |
| 資源タイプ |
|
|
資源タイプ識別子 |
http://purl.org/coar/resource_type/c_18gh |
|
資源タイプ |
technical report |
| 著者所属 |
|
|
|
Osaka University |
| 著者所属 |
|
|
|
Osaka Electro-Communication University |
| 著者所属 |
|
|
|
Osaka University |
| 著者所属(英) |
|
|
|
en |
|
|
Osaka University |
| 著者所属(英) |
|
|
|
en |
|
|
Osaka Electro-Communication University |
| 著者所属(英) |
|
|
|
en |
|
|
Osaka University |
| 著者名 |
Zhenjiao, Wang
Yoshimichi, Ito
Noboru, Babaguchi
|
| 著者名(英) |
Zhenjiao, Wang
Yoshimichi, Ito
Noboru, Babaguchi
|
| 論文抄録 |
|
|
内容記述タイプ |
Other |
|
内容記述 |
This paper proposes several methods for computing upper and lower bounds of L2 error of multiview triangulation. The multiview triangulation is to find the point that minimizes the sum of reprojection errors calculated by points on the image plane observed by multiple cameras. We first show the L2 optimization for multiview triangulation is reduced to the ones with nonconvex matrix inequality constraints, which are hard to solve. By relaxing the nonconvex matrix inequality constraints, we derive conditions for computing lower bounds of L2 optimal error. The conditions are represented by linear matrix inequalities (LMI), and they are easy to solve. On the other hand, by tightening the nonconvex matrix inequality constraints, we derive conditions for computing upper bounds of L2 optimal error, which are also represented by LMI. These methods are easily implemented by MATLAB using tools for LMI such as SeDuMi and YALMIP. The proposed methods are evaluated through numerical examples. |
| 論文抄録(英) |
|
|
内容記述タイプ |
Other |
|
内容記述 |
This paper proposes several methods for computing upper and lower bounds of L2 error of multiview triangulation. The multiview triangulation is to find the point that minimizes the sum of reprojection errors calculated by points on the image plane observed by multiple cameras. We first show the L2 optimization for multiview triangulation is reduced to the ones with nonconvex matrix inequality constraints, which are hard to solve. By relaxing the nonconvex matrix inequality constraints, we derive conditions for computing lower bounds of L2 optimal error. The conditions are represented by linear matrix inequalities (LMI), and they are easy to solve. On the other hand, by tightening the nonconvex matrix inequality constraints, we derive conditions for computing upper bounds of L2 optimal error, which are also represented by LMI. These methods are easily implemented by MATLAB using tools for LMI such as SeDuMi and YALMIP. The proposed methods are evaluated through numerical examples. |
| 書誌レコードID |
|
|
収録物識別子タイプ |
NCID |
|
収録物識別子 |
AA11131797 |
| 書誌情報 |
研究報告コンピュータビジョンとイメージメディア（CVIM）
巻 2016-CVIM-201,
号 1,
p. 1-6,
発行日 2016-02-25
|
| ISSN |
|
|
収録物識別子タイプ |
ISSN |
|
収録物識別子 |
2188-8701 |
| Notice |
|
|
|
SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc. |
| 出版者 |
|
|
言語 |
ja |
|
出版者 |
情報処理学会 |
IPSJ-CVIM16201001.pdf (617.3 kB)
