A new real-time object tracking algorithm for effect and efficiency optimization

A novel real-time object tracking algorithm for effect and efficiency optimization is proposed. In this paper, the rectangle and high-dimensional features at different scales of positive and negative samples are extracted. Then a measurement matrix is constructed to map high-dimensional features to lower-dimensional image space using a prior knowledge of sparse video frame. High features are persisted and the dimension of them is reduced, therefore, the storage space is reduced and calculation speed is raised. A naïve Bayes classifier is used which is updated with a prior knowledge of former frame. Lower-dimensional features are inputted to the updated classifiers to develop the Maximal likelihood estimation. Experimental results show that the proposed algorithm can handle occlusion efficiently, and be robust to pose and illumination variations.