Maneuvering object tracking under large-area occlusive condition using mean shift embedded IMM filter

Visual tracking is the act of consistently locating a desired feature in each image of a given input sequence and it is typically complicated by several factors such as noise of sensors, motion in the scene, motion on the part of the observer and real-time constraints. In this paper, a novel visual tracking approach is proposed by jointly using mean shift and IMM, which can effectively deal with the visual tracking problem. The proposed approach can also resolve the target large-area occlusions. Based on the experimental results provided, it can be concluded that our proposed approach is rational and effective.