The problem of deformable objects in complex, cluttered
environments is addressed. This approach, called the Cluster
Tracker, concentrates on sectioning and tracking small features
rather than the whole image as more conventional trackers do. Results
indicate that the Cluster Tracker exhibits higher tracking reliability
in heavily cluttered environments due to its ability to maintain
track lock on objects even when they are partially obscured. The
core of this method is based on an elastic matrix framework that
supports a flexible structured model of the target. This approach
allows the tracker to follow deformations of the target’s body shape
and to estimate the location of occluded features.
Andrew Cilia,
"Object tracking using cluster of elastically linked feature trackers," Journal of Electronic Imaging 17(2), 023019 (1 April 2008). https://doi.org/10.1117/1.2931494