A new algorithm is presented for feature point based motion tracking in lon
g image sequences. Dynamic scenes with multiple, independently moving objec
ts are considered in which feature points may temporarily disappear, enter
and leave the view field. This situation is typical for surveillance and sc
ene monitoring applications.
Most of the existing approaches to feature point tracking have limited capa
bilities in handling incomplete trajectories, especially when the number of
points and their speeds are large, and trajectory ambiguities are frequent
. The proposed algorithm was designed to efficiently resolve these ambiguit
ies. Correspondences between moving points are established in a competitive
linking process that develops as the trajectories grow. Appearing and disa
ppearing points are treated in a natural way as the points that do not link
.
The proposed algorithm compares favorably to efficient alternative algorith
ms selected and tested in a performance evaluation study.