Title :
Initialization and System Modeling in 3-D Pose Tracking
Author :
Kragic, Danica ; Kyrki, Ville
Author_Institution :
Computational Vision & Active Perception, KTH, Stockholm
Abstract :
Initialization and choice of adequate motion models are two important but seldom discussed problems in 3D model-based pose (position and orientation) tracking. In this paper, we propose an automatic initialization approach suitable for textured objects. In addition, we define, study and experimentally evaluate three motion models commonly used in visual servoing and augmented reality
Keywords :
Kalman filters; augmented reality; image motion analysis; image texture; stereo image processing; tracking; 3D pose tracking; augmented reality; automatic initialization approach; motion model; motion models; orientation tracking; position tracking; system modeling; textured objects; visual servoing; Augmented reality; Cameras; Computer vision; Histograms; Information processing; Laboratories; Modeling; Predictive models; Tracking; Visual servoing;
Conference_Titel :
Pattern Recognition, 2006. ICPR 2006. 18th International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
0-7695-2521-0
DOI :
10.1109/ICPR.2006.712
