Contour Based Object Tracking
High level vision tasks for video processing, such as recognition and understanding, require tracking of complete objects. In this paper, we propose a contour tracking method for video acquired using mobile cameras, which can track the complete objects. The proposed method can track multiple objects, adapt to changing visual features, and handle occlusions. Our approach has two major components related to visual features and object shape. Visual features (color, texture) are modeled by semi-parametric models, where the mixing parameters are computed using independent opinion polling. The shape prior, which is used to fill missing observations for occluded objects, is modeled using parametric models. We formulate the contour energy as a variational calculus problem, which results in a system of nonlinear partial integro-differential equations of order one. The energy is minimized in the gradient descent direction evaluated in the contour vicinity defined by a band. In this regard, it can be viewed as the generalization of formerly proposed contour based methods. The performance of the proposed method is demonstrated on real sequences with and without complete object occlusions.
A. Yilmaz, X. Li and M. Shah, Contour Based Object Tracking with Occlusion Handling in Video Acquired Using Mobile Cameras, IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI), Vol.26, No. 11, 2004.
A. Yilmaz and M. Shah, Contour Based Object Tracking Using Level Sets, Proceedings of Asian Conf. on Computer Vision (ACCV), South Korea, 2004.