Object Tracking Using Multiple Features and Adaptive Model Updating
Qingyong Hu,Yulan Guo,Zaiping Lin,Wei An,Hongwei Cheng
DOI: https://doi.org/10.1109/tim.2017.2729378
IF: 5.6
2017-01-01
IEEE Transactions on Instrumentation and Measurement
Abstract:Correlation filter-based tracking methods have been intensively investigated for their high efficiency and robustness. However, a single feature-based tracker cannot adapt to challenging situations, such as severe deformation, rotation, and illumination variations. Besides, a simple linear interpolation-based model updating mechanism is prone to model degradation, and consequently tracker drifting. In this paper, a 2-D location filter is combined with a 1-D scale filter to jointly estimate the state of object under tracking, and three complementary features are integrated to further enhance the overall tracking performance. Besides, we define a penalty factor for adaptive model updating, to achieve a balance between stability and flexibility, especially when the object is under occlusion. Extensive experiments have been conducted on four large-scale data sets, namely, the object tracking benchmark, VOT15, Temple-Color128, and the UAV123 tracking benchmark. Quantitative and qualitative results show that our proposed tracker achieves promising results in terms of tracking accuracy, robustness, and speed as compared with other popular trackers, and is highly suitable for real-time applications, such as unmanned aerial vehicles. It outperforms the state-of-the-art methods under different nuisances, including scale variation, deformation, occlusion, rotation, and out-of-view.