Visual tracking using D2-clustering and particle filter

Since tracking algorithms should be robust with respect to appearance changes, online algorithms has been investigated recently instead of offline ones which has shown an acceptable performance in controlled environments. The most challenging issue in online algorithms is updating of the model causing tracking failure because of introducing small errors in each update and disturbing the appearance model (drift). In this paper, we propose an online generative tracking algorithm in order to overcome the challenges such as occlusion, object shape changes, and illumination variations. In each frame, color distribution of target candidates is obtained and the candidate having the lowest distance to the object distribution is considered as the object. In addition, in our work, the particle filter structure is used in which the samples are weighted proportional to their distance to the model. The model which is a color distribution is updated using D2-clustring algorithm. The most distinctive features of our algorithm are: 1) Updating the model using D2-clustering, 2) Avoiding drifting by using the color distribution of the target in the first and last frame, and 3) Detecting of occlusion by considering distance between the model and the best candidate. Experimental results show that our tracker outperforms other algorithms in videos containing challenging scenarios.