Robust Object Tracking in Crowded Scenes Based on the Undecimated Wavelet Features and Particle Filter
Hamed Moradipour, Hossein Ashtiani, Amir Aliabadian · 2010
Abstract: A Scale Invariant Feature Transform (SIFT) based on particle filter algorithm is presented for object tracking. We propose a new algorithm for object tracking in crowded video scenes by exploiting the properties of Undecimated Wavelet Packet Transform (UWPT) and particle filter. SIFT features are used to correspond the region of interests across frames. Meanwhile, feature vectors generated via the coefficients of the UWPT is applied to conduct similarity search that is based on particle filter. The advantage of using structural similarity index UWPT domain is that it allows spatial translations, rotations and scaling changes. Experimental results show that the proposed algorithm has good performance for object tracking in noisy crowded scenes on stairs, in airports, or at train stations in the presence of object translation, rotation, small scaling and occlusion. Key words: Object tracking • sift feature • undecimated wavelet transform • particle filter