Object of interest tracking based on visual saliency and feature points matching

To solve the problems of traditional object tracking algorithms which need to select the object manually and the tracking of the interested object is unstable, a new interested object tracking algorithm based on visual saliency and feature points matching is proposed in this paper by combining the human visual attention mechanism with multi-scale local characteristic theory. The algorithm mainly contains two parts: the object of interest detection and object tracking. Firstly, the object of interest to be tracked is detected based on visual saliency in the initial frame; then, the detected object is tracked by SIFT bidirectional matching. The experimental results demonstrate the effectiveness and the robustness of the proposed method. Compared with the traditional techniques, the proposed method is in better correspondence with the response of the human visual system and the perception mechanism.