Robust matching in real-time 3D model-based tracking

In this paper, a new model-based tracking algorithm is proposed for real time performance. The matching process includes two aspects of: feature extraction using local minimum energy; and global matching of a known 3D model against the projected features. The algorithm is robust to change in lighting and backgrounds. The small motion hypothesis is used for fitting of feature energy which is defined as the negative absolute value of the edge strength. An autoregressive AR(1) model is employed for detecting incorrect matches in terms of the feature energy.