Combining the spatial and temporal eigen-space for visual tracking

Visual tracking is an important research topic in computer vision community. Most subspace based tracking algorithms focus on the time correlation between the image observations of the object, but the spatial layout information of the object is ignored. This paper proposes a robust visual tracking algorithm which effectively combines the spatial and temporal eigen-space of the object. In order to captures the variations of object appearance, an incremental updating strategy is developed to update the eigen-space and mean of the object. Experimental results demonstrate that, compared with the state-of-the-art subspace based tracking algorithms, the proposed tracking algorithm is more robust and effective.