Planar object tracking plays an important role in computer vision and relatedfields. While several benchmarks have been constructed for evaluatingstate-of-the-art algorithms, there is a lack of video sequences captured in thewild rather than in constrained laboratory environment. In this paper, wepresent a carefully designed planar object tracking benchmark containing 210videos of 30 planar objects sampled in the natural environment. In particular,for each object, we shoot seven videos involving various challenging factors,namely scale change, rotation, perspective distortion, motion blur, occlusion,out-of-view, and unconstrained. The ground truth is carefully annotatedsemi-manually to ensure the quality. Moreover, eleven state-of-the-artalgorithms are evaluated on the benchmark using two evaluation metrics, withdetailed analysis provided for the evaluation results. We expect the proposedbenchmark to benefit future studies on planar object tracking.
展开▼
