A Robust Object Tracking Method Using Structural Similarity in Daubechies Complex Wavelet Domain

Abstract

Many of the existing algorithms for object tracking that are based on spatial domain features, fail in the presence of illumination variation or change in appearance or pose or in the presence of noise. To overcome these problems, in this paper, we have proposed a new method of object tracking using structural similarity index in complex wavelet transform domain, which is approximately shift-invariant. The reference object in the initial frame is modeled by a feature vector in terms of the coefficients of Daubechies complex wavelet transform. A similarity measure based on structural similarity index is used to find the object in the current frame. The advantage of using structural similarity index in complex wavelet domain is that it allows small spatial translations, rotations and scaling changes, which are depicted in fig. 1. Experimental results illustrate that the proposed algorithm has good performance in noisy video with significant variations in object's pose and illumination. The search for the candidate subframe is made fast by using the motion prediction algorithm.