Title :
Tracking objects using density matching and shape priors
Author :
Zhang, Tao ; Freedman, Daniel
Author_Institution :
Comput. Sci. Dept., Rensselaer Polytech. Inst., Troy, NY, USA
Abstract :
We present a novel method for tracking objects by combining density matching with shape priors. Density matching is a tracking method which operates by maximizing the Bhattacharyya similarity measure between the photometric distribution from an estimated image region and a model photometric distribution. Such trackers can be expressed as PDE-based curve evolutions, which can be implemented using level sets. Shape priors can be combined with this level-set implementation of density matching by representing the shape priors as a series of level sets; a variational approach allows for a natural, parametrization-independent shape term to be derived. Experimental results on real image sequences are shown.
Keywords :
Kalman filters; computer vision; image representation; image segmentation; image sequences; object detection; partial differential equations; splines (mathematics); B-splines; Bhattacharyya similarity; Euclidean distance; Euclidean similarity transformation; Kullbach-Leibler distance; Mahalanobis distance; PDE-based curve evolutions; active contours; computer vision; density matching; differential equations; image region; image sequences; level set method; object tracking; photometric distribution; segmentation method; shape energy; shape priors; Active contours; Biomedical imaging; Computer science; Computer vision; Density measurement; Image segmentation; Image sequences; Level set; Photometry; Shape;
Conference_Titel :
Computer Vision, 2003. Proceedings. Ninth IEEE International Conference on
Conference_Location :
Nice, France
Print_ISBN :
0-7695-1950-4
DOI :
10.1109/ICCV.2003.1238466
