Details
| Originalsprache | Englisch |
|---|---|
| Titel des Sammelwerks | Algebraic Frames for the Perception-Action Cycle - 2nd International Workshop, AFPAC 2000, Proceedings |
| Herausgeber (Verlag) | Springer Verlag |
| Seiten | 339-348 |
| Seitenumfang | 10 |
| ISBN (elektronisch) | 3540410139, 9783540410133 |
| Publikationsstatus | Veröffentlicht - 2000 |
| Extern publiziert | Ja |
| Veranstaltung | 2nd International Workshop on Algebraic Frames for the Perception-Action Cycle, AFPAC 2000 - Kiel, Deutschland Dauer: 10 Sept. 2000 → 11 Sept. 2000 |
Publikationsreihe
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Band | 1888 |
| ISSN (Print) | 0302-9743 |
| ISSN (elektronisch) | 1611-3349 |
Abstract
The paper develops three extended Kalman filters (EKF) for 2D-3D pose estimation. The measurement models are based on three constraints which are constructed by geometric algebra. The dynamic measurements for these EKF are either points or lines. The real monocular vision experiments show that the results of EKFs perform more stable than that of LMS method.
ASJC Scopus Sachgebiete
- Mathematik (insg.)
- Theoretische Informatik
- Informatik (insg.)
Zitieren
- Standard
- Harvard
- Apa
- Vancouver
- BibTex
- RIS
Algebraic Frames for the Perception-Action Cycle - 2nd International Workshop, AFPAC 2000, Proceedings. Springer Verlag, 2000. S. 339-348 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Band 1888).
Publikation: Beitrag in Buch/Bericht/Sammelwerk/Konferenzband › Aufsatz in Konferenzband › Forschung › Peer-Review
}
TY - GEN
T1 - Extended kalman filter design for motion estimation by point and line observations
AU - Zhang, Yiwen
AU - Rosenhahn, Bodo
AU - Sommer, Gerald
PY - 2000
Y1 - 2000
N2 - The paper develops three extended Kalman filters (EKF) for 2D-3D pose estimation. The measurement models are based on three constraints which are constructed by geometric algebra. The dynamic measurements for these EKF are either points or lines. The real monocular vision experiments show that the results of EKFs perform more stable than that of LMS method.
AB - The paper develops three extended Kalman filters (EKF) for 2D-3D pose estimation. The measurement models are based on three constraints which are constructed by geometric algebra. The dynamic measurements for these EKF are either points or lines. The real monocular vision experiments show that the results of EKFs perform more stable than that of LMS method.
UR - http://www.scopus.com/inward/record.url?scp=33846645600&partnerID=8YFLogxK
U2 - 10.1007/10722492_27
DO - 10.1007/10722492_27
M3 - Conference contribution
AN - SCOPUS:33846645600
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 339
EP - 348
BT - Algebraic Frames for the Perception-Action Cycle - 2nd International Workshop, AFPAC 2000, Proceedings
PB - Springer Verlag
T2 - 2nd International Workshop on Algebraic Frames for the Perception-Action Cycle, AFPAC 2000
Y2 - 10 September 2000 through 11 September 2000
ER -