Graph-Embedding-Based Learning for Robust Object Tracking

Object tracking is viewed as a two-class "one-versus-rest" classification problem, in which the sample distribution of the target over a short period of time is approximately Gaussian while the background samples are often multimodal. Based on these special properties, we propose a graph-embedding-based learning method, in which the topology structures of graphs are carefully designed to reflect the properties of the sample distributions. This method can simultaneously learn the subspace of the target and its local discriminative structure against the background. Moreover, a heuristic negative sample selection scheme is adopted to make the classification more effective. In applications to tracking, the graph-embedding-based learning is incorporated into a Bayesian inference framework cascaded with hierarchical motion estimation, which significantly improves the accuracy and efficiency of the localization. Furthermore, an incremental updating technique for the graphs is developed to capture the changes in both appearance and illumination. Experimental results demonstrate that, compared with the two state-of-the-art methods, the proposed tracking algorithm is more efficient and effective, particularly in dynamically changing and cluttered scenes.