| ECVision indexed and annotated bibliography of cognitive computer vision publications This bibliography was created by Hilary Buxton and Benoit Gaillard, University of Sussex, as part of ECVision Specific Action 8-1 The complete text version of this BibTeX file is available here: ECVision_bibliography.bib
J. Biber and M. Vincze and M. Ayromlou and W. Ponweiser
ABSTRACT
Ellipse arcs are typical features of man-made objects and tracking of ellipse arcs is needed in robotics applications where partial occlusion is a standard problem that should be handled. The diĘculty in tracking of ellipse arcs is to robustly locate the edge and to t the arc, that is, the visible segment of the ellipse. Robust edge detection is obtained using an improved version of the approach using Edge Projected Integration of Cues (EPIC). It is shown how EPIC highly reduces the number of outliers and that a probabilistic tting method obtains robust ellipse estimates for subtended angles larger than a half circle. The ellipse arc tracker is able to automatically detect occlusion. From the ellipses and a related object model the pose of the part can be reconstructed and this signal can be used for motion control. |
|
| Site generated on Friday, 06 January 2006
|