Kernel Circulant Object Tracking Based on Illumination Invariant Features
In order to improve accuracy and robustness of object tracking, and to resolve the problem that kernel circulant object tracking is easily affected by illumination changes, this paper presents a kernel circulant object tracking method based on illumination invariant features. Firstly, the locality sensitive histograms of input image are calculated and illumination invariant features are extracted. Then, the precise position of tracking object is obtained by the responded confidence image, which can be quickly computed in frequency domain between input image and template image. Tests of many video sequences demonstrate that the proposed method can track the target accurately and effectively. Comparing to the conditional kernel circulant tracking algorithm, the tracking error using our method decreases 24 pixels and the tracking precision increases 28%. As a result, the proposed algorithm has the better robustness under abruptly variant illumination and pose.
KeywordsObject tracking Locality sensitive histograms Illumination invariant features Kernel circulant algorithm
This research was supported by National Natural Science Foundation of China (No. 61771230).
- 1.Wang, X., Hua, G., Han, T.X.: Discriminative tracking by metric learning. In: Proceedings of European Conference on Computer Vision, Crete, Greece, pp. 200–214 (2010)Google Scholar
- 4.Zhang, K., Zhang, L., Yang, M.H.: Real-time compressive tracking. In: ECCV, pp. 866–879 (2012)Google Scholar
- 6.Henriques, J.F., Caseiro, R., Martins, P., et al.: Exploiting the circulant structure of tracking-by-detection with kernels. In: ECCV, pp. 702–715 (2012)Google Scholar
- 9.He, S.F., Yang, Q.X., Rynson, W.H.L., et al.: Visual tracking via locality sensitive histograms. In: CVPR, pp. 2427–2434 (2013)Google Scholar
- 11.Tian, H., Ju, Y.F., Meng, F.K., Li, F.F.: Improved tracking algorithm with background-weighted histogram. J. Image Graph. 20(1), 0072–0084 (2015)Google Scholar
- 12.Adam, A., Rivlin, E., Shimshoni, I.: Robust fragments-based tracking using the integral histogram. In: CVPR, pp. 798–805 (2006)Google Scholar