Long-Term Tracking Algorithm Based on Kernelized Correlation Filter
KCF (Kernelized Correlation Filter) is a classical tracking algorithm based on correlation filter, which has good performance in short-term tracking. But when the object is partially or fully occluded, or disappeared in the view, KCF doesn’t work well. In this paper, a long-term tracking algorithm based on KCF is proposed. HOG (Histogram of Oriented Gradient) and LAB three-channel color information are employed to represent the object, and a re-detection module is added into the KCF tracking procedure. The peak ratio is introduced to control the start of the re-detection module, and a correlation filter model based on SURF feature points is re-learned to continuously track the occluded object. Experimental results on OTB dataset show that our algorithm has higher tracking accuracy than other five trackers, and is suitable for long-term tracking.
KeywordsLong-term object tracking Correlation filter Re-detection
The work is sponsored by the Shaanxi International Cooperation Exchange Funded Projects (2017KW-013, 2017KW-016), Graduate Creative Funds of Xi’an University of Posts and Telecommunications (CXJJ2017004).
- 4.Bolme, D.S., Beveridge, J.R., Draper, B.A.: Visual object tracking using adaptive correlation filters. In: Computer Vision and Pattern Recognition, pp. 2544–2550. IEEE (2010). https://doi.org/10.1109/cvpr.2010.5539960
- 8.Danelljan, M., Häger, G., Fahad, S.K., et al.: Accurate scale estimation for robust visual tracking. In: British Machine Vision Conference, pp. 1–5. BMVA Press (2014). https://doi.org/10.5244/c.28.65
- 13.Ma, C., Yang, X., Zhang, C., et al.: Long-term correlation tracking. In: Computer Vision and Pattern Recognition, pp. 5388–5396. IEEE (2015). https://doi.org/10.1109/cvpr.2015.7299177
- 14.Zhao, G.P., Shen, Y.P., Wang. J.Y.: Adaptive feature fusion object tracking based on circulant structure with kernel. Acta Optica Sinica 37, 0815001. https://doi.org/10.3788/aos201737
- 15.Wu, Y., Lim, J., Yang, M.H.: Online object tracking: a benchmark. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 2411–2418. IEEE (2013). https://doi.org/10.1109/cvpr.2013.312
- 16.Danelljan, M., Khan, F.S., Felsberg, M., et al.: Adaptive color attributes for real-time visual tracking. In: Computer Vision and Pattern Recognition, pp. 1090–1097. IEEE (2014). https://doi.org/10.1109/cvpr.2014.143