Contour Tracking for Non-Rigid Objects by Using Supervised Level Set Model

Abstract

In this paper we can track the objects according to their shape. We can track the non-rigid objects by using supervised level set model (SLSM).In existing method we use the bounding box method track the object. The proposed method extracts the accurate contours of the target as tracking output, which achieves better description of the non-rigid objects while reduces background pollution to the target model. The SLSM can ensure a more accurate convergence to the exact targets in tracking applications. In particular, we firstly construct the appearance model for the target in an Online boosting manner due to its strong discriminative power between the object and the background. We firstly describe the proposed mechanism of two-phase SLSM for single target tracking, and then give its generalized multi-phase version for dealing with multi-target tracking cases. In SLSM method we apply the contour on non-rigid objects to track the objects exactly. Contour Based Object Tracking is an efficient method for tracking Objects, which can detect multiple objects simultaneously. Contour based object tracking can track the Objects in both images and videos. In this method we will find the contours of the objects, then using this contours we will track the Objects. Finding the contours is the most important task in the contour based object tracking.