A method of reactive zoom control from uncertainty in tracking
Computer Vision and Image Understanding
Comparing active vision models
Image and Vision Computing
3D Target Scale Estimation and Target Feature Separation for Size Preserving Tracking in PTZ Video
International Journal of Computer Vision
Dynamic view planning by effective particles for three-dimensional tracking
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics - Special issue on human computing
Multi-step multi-camera view planning for real-time visual object tracking
DAGM'06 Proceedings of the 28th conference on Pattern Recognition
Cognitive visual tracking and camera control
Computer Vision and Image Understanding
Camera selection for tracking in distributed smart camera networks
ACM Transactions on Sensor Networks (TOSN)
Online control of active camera networks for computer vision tasks
ACM Transactions on Sensor Networks (TOSN)
| Hi-index | 0.00 |
Active object tracking, for example, in surveillance tasks, becomesmore and more important these days. Besides the tracking algorithmsthemselves methodologies have to be developed for reasonable activecontrol of the degrees of freedom of all involved cameras. In thispaper we present an information theoretic approach that allows theoptimal selection of the focal lengths of two cameras during active3-D object tracking. The selection is based on the uncertainty inthe 3-D estimation. This allows us to resolve the trade-off betweensmall and large focal length: in the former case, the chance isincreased to keep the object in the field of view of the cameras.In the latter one, 3-D estimation becomes more reliable. Also, moredetails are provided, for example for recognizing theobjects.Beyond a rigorous mathematical framework we presentreal-time experiments demonstrating that we gain an improvementin3-D trajectory estimation by up to 42% in comparison with trackingusing a fixed focal length.