An appearance model adaptable to changes in object appearance is critical in visual object tracking. In\udthis paper, we treat an image patch as a 2-order tensor which preserves the original image structure. We design\udtwo graphs for characterizing the intrinsic local geometrical structure of the tensor samples of the object and the\udbackground. Graph embedding is used to reduce the dimensions of the tensors while preserving the structure of\udthe graphs. Then, a discriminant embedding space is constructed. We prove two propositions for finding the\udtransformation matrices which are used to map the original tensor samples to the tensor-based graph embedding\udspace. In order to encode more discriminant information in the embedding space, we propose a transfer-learningbased\udsemi-supervised strategy to iteratively adjust the embedding space into which discriminative information\udobtained from earlier times is transferred. We apply the proposed semi-supervised tensor-based graph\udembedding learning algorithm to visual tracking. The new tracking algorithm captures an object’s appearance\udcharacteristics during tracking and uses a particle filter to estimate the optimal object state. Experimental results\udon the CVPR 2013 benchmark dataset demonstrate the effectiveness of the proposed tracking algorithm.
展开▼
