Title :
Tracking feature points: a new algorithm
Author :
Chetverikov, Dmitry ; Verestoy, Judit
Author_Institution :
Comput. & Autom. Res. Inst., Budapest, Hungary
Abstract :
A new algorithm is presented for feature point based motion tracking in long image sequences. Dynamic scenes with multiple, independently moving objects are considered in which feature points may temporarily disappear enter and leave the view field. The existing approaches to feature point tracking have limited capabilities 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 ambiguities
Keywords :
feature extraction; image sequences; motion estimation; tracking; dynamic scenes; feature point based motion tracking; incomplete trajectories; long image sequences; multiple independently moving objects; trajectory ambiguities; Algorithm design and analysis; Automation; Electrical capacitance tomography; Image motion analysis; Image sequence analysis; Image sequences; Layout; Motion analysis; Pattern analysis; Tracking;
Conference_Titel :
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-8186-8512-3
DOI :
10.1109/ICPR.1998.711973