Robust visual tracking based on L1 expanded template
This data was imported from Scopus:
Authors: Cheng, D., Zhang, Y., Tian, F., Shi, D. and Liu, X.
Journal: Proceedings of 2017 International Conference on Machine Learning and Cybernetics, ICMLC 2017
© 2017 IEEE. Most video tracking algorithms including L1 tracker often fail to track correctly under adverse conditions such as object occlusion, disappearance, etc. To address this issue, we propose an improved L1 tracker algorithm called Tracker-2, based on what we call the expanded template which includes the reference template and trail template. The reference template keeps the original features of the target and prevents errors from being introduced by false tracking results with the template update, which leads to the deviation of the target. The trail template records the trail tracking results to avoid massive use of trivial templates which may result in the false detection of occlusion. The experimental results on a number of standard data sets have proved that our Tracker-2 approach is able to deal with the occlusion problem effectively while maintaining the advantages of L1 tracker.