Confidence Based Hierarchical Online Ensemble Visual Tracking

Abstract:

In order to adapt to the target appearance changes during visual tracking, feature model needs to be updated by online learning. However, online adaptive methods suffer from the drifting problem: error data are used for updating and will finally lead to tracking failure. In this paper, we propose a novel hierarchical online ensemble tracking method. Optical flow tracker is employed to predict the rough location. Online learning classifier is employed to detect the target. Template match is used to filter the data for updating. All these parts are combined together hierarchically by their confidence to ensure the stability of online learning and tracking. In contrast to the individual online learning and semi-supervised online learning, our method show more adaptability and stability. We demonstrate the performance on several public video sequences, which shows the improvement of our method over other online tracking approaches.

• Periodical: Advanced Materials Research (Volumes 945-949)
• Pages: 1794-1800
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.945-949.1794
• Online since: June 2014
• Authors: Wen Hua He*, Zhi Jing Liu, Jian Ming Qu
• Keywords: Ensemble Tracking, Real Time, Semi-Supervised Learning, Visual Tracking

References

[1] Grabner H, Bischof H, Online boosting and visions, Computer Vision and Pattern Recognition, Computer Society Conference on. IEEE, 2006, 1: 260-267.

DOI: 10.1109/cvpr.2006.215

[2] Grabner H, Leistner C, Bischof H, Semi-supervised online boosting for robust tracking, Computer Vision–ECCV, Springer Berlin Heidelberg, 2008: 234-247.

DOI: 10.1007/978-3-540-88682-2_19

[3] Babenko B, Yang M H, Belongie S. Visual tracking with online multiple instance learning, Computer Vision and Pattern Recognition , IEEE Conference on. IEEE, 2009: 983-990.

DOI: 10.1109/cvpr.2009.5206737

[4] Oza N C, Russell S, Online ensemble learning, AAAI/IAAI, 2000: 1109.

[5] Adam A, Rivlin E, Shimshoni I, Robust fragments-based tracking using the integral histogram, Computer Vision and Pattern Recognition, Computer Society Conference on. IEEE, 2006, 1: 798-805.

DOI: 10.1109/cvpr.2006.256

[6] Avidan S, Pattern Analysis and Machine Intelligence, IEEE Transactions, 2007, 29(2): 261-271.

[7] Hare S, Saffari A, Torr P H S, Struck: Structured output tracking with kernels, ICCV, IEEE International Conference on. IEEE, 2011: 263-270.

DOI: 10.1109/iccv.2011.6126251

[8] Kalal Z, Mikolajczyk K, Matas J, Tracking-learning-detection, Pattern Analysis and Machine Intelligence, IEEE Transactions on, 2012, 34(7): 1409-1422.

DOI: 10.1109/tpami.2011.239

[9] Santner J, Leistner C, Saffari A, Prost: Parallel robust online simple tracking, Computer Vision and Pattern Recognition, IEEE Conference on. IEEE, 2010: 723-730.

DOI: 10.1109/cvpr.2010.5540145

[10] Kalal Z, Mikolajczyk K, Matas J, Forward-backward error: Automatic detection of tracking failures, ICPR, 2010 20th International Conference on. IEEE, 2010: 2756-2759.

DOI: 10.1109/icpr.2010.675

[11] Bouguet J Y, Pyramidal implementation of the affine lucas kanade feature tracker description of the algorithm, Intel Corporation, 2001, 2: 3.

[12] Wu Y, Lim J, Yang M H, Online object tracking: A benchmark, Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. IEEE, 2013: 2411-2418.

DOI: 10.1109/cvpr.2013.312