Robust Visual Tracking with SIFT Features and Fragments Based on Particle Swarm Optimization

详细信息    查看全文

• 作者：Xu Cheng (1)
  Nijun Li (1)
  Suofei Zhang (1)
  Zhenyang Wu (1)
  
• 关键词：Visual tracking ; Appearance variation ; Particle swarm optimization ; RANSAC ; SIFT
• 刊名：Circuits, Systems, and Signal Processing
• 出版年：2014
• 出版时间：May 2014
• 年：2014
• 卷：33
• 期：5
• 页码：1507-1526
• 全文大小：
• 参考文献：1. A. Adam, E. Rivlin, I. Shimshoni, Robust fragments-based tracking using the integral histogram, in / Proc. Int. Conf. Comp. Vis. Pattern Recognition (2006), pp. 798鈥?05
  2. M.S. Arulampalam, S. Maskell, N. Gordon, T. Clapp, A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. IEEE Trans. Signal Process. 50(2), 174鈥?88 (2002) CrossRef
  3. S. Avidan, Ensemble tracking. IEEE Trans. Pattern Anal. Mach. Intell. 29(2), 261鈥?71 (2007) CrossRef
  4. B. Babenko, M. Yang, S. Belongie, Visual tracking with online multiple instance learning. IEEE Trans. Pattern Anal. Mach. Intell. 33(8), 1619鈥?632 (2010) CrossRef
  5. S.T. Birchfield, S. Rangarajan, Spatiograms versus histograms for region-based tracking, in / Proc. Int. Conf. Comp. Vis. Pattern Recognition (2005), pp. 1158鈥?163
  6. CAVIAR datasets, http://groups.inf.ed.ac.uk/vision/caviar/caviardata1/
  7. R. Collins, Y. Liu, M. Leordeanu, On-line selection of discriminative tracking features. IEEE Trans. Pattern Anal. Mach. Intell. 27(10), 1631鈥?643 (2005) CrossRef
  8. D. Comaniciu, P. Meer, Mean shift: a robust approach toward feature space analysis. IEEE Trans. Pattern Anal. Mach. Intell. 24(5), 603鈥?19 (2002) CrossRef
  9. D. Comaniciu, V. Ramesh, P. Meer, Kernel-based object tracking. IEEE Trans. Pattern Anal. Mach. Intell. 25(5), 564鈥?77 (2003) CrossRef
  10. E. Erdem, S. Dubuisson, I. Bloch, Fragments based tracking with adaptive cue integration. Comput. Vis. Image Underst. 116(7), 827鈥?41 (2012) CrossRef
  11. B. Erkmen, R. Vural, N. Kahraman et al., A mixed mode neural network circuitry for object recognition application. Circuits Syst. Signal Process. 32(1), 29鈥?6 (2013) CrossRef
  12. M.A. Fischer, R.C. Bolles, Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM 24(6), 381鈥?95 (1981) CrossRef
  13. K. Fukunaga, L. Hostetler, The estimation of the gradient of a density function, with applications in pattern recognition. IEEE Trans. Inf. Theory 21(1), 32鈥?0 (1975) CrossRef
  14. H. Grabner, M. Grabner, H. Bischof, Real-time tracking via on-line boosting, in / Proc. British Machine Vis. Conf. (2006), pp. 47鈥?6
  15. W. Heng, K. Ngan, Long transition analysis for digital video sequences. Circuits Syst. Signal Process. 20(2), 113鈥?41 (2001) CrossRef
  16. M. Isard, A. Blake, Condensation-conditional density propagation for visual tracking. Int. J. Comput. Vis. 29(1), 5鈥?8 (1998) CrossRef
  17. K. James, R. Eberhart, Particle swarm optimization, in / Proceedings of IEEE International Conference on Neural Networks (1995), pp. 1942鈥?948
  18. J. Kwon, K.M. Lee, Visual tracking decomposition, in / Proc. Int. Conf. Comp. Vis. Pattern Recognition (2010), pp. 1269鈥?276
  19. D. Lowe, Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60(2), 91鈥?10 (2004) CrossRef
  20. X. Mei, H. Ling, Robust visual tracking using L1 minimization, in / Proc. IEEE Int. Conf. Comp. Vis. (2009), pp. 1436鈥?443
  21. K. Nummiaro, E. Koller-Meier, L.V. Gool, An adaptive color-based particle filter. Image Vis. Comput. 21(1), 99鈥?10 (2003) CrossRef
  22. K. Okuma, A. Taleghani, O. Freitas, A boosted particle filter: multitarget detection and tracking, in / The European Conference Computer Vision (2004), pp. 28鈥?9
  23. D. Ross, J. Lim, R.S. Lin, M.H. Yang, Incremental learning for robust visual tracking. Int. J. Comput. Vis. 77(1), 125鈥?41 (2008) CrossRef
  24. Y. Rubner, C. Tomasi, L.J. Guibas, The earth mover鈥檚 distance as a metric for image retrieval. Int. J. Comput. Vis. 40(2), 99鈥?21 (2000) CrossRef
  25. M. Thida, P. Remagnino, H. Eng, A particle swarm optimization approach for multi-objects tracking in crowded scene, in / IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops (2009), pp. 1209鈥?215 CrossRef
  26. M. Thida, H. Eng, D.N. Monekosso, P. Remagnino, A particle swarm optimization algorithm with interactive swarms for tracking. Appl. Soft Comput. 13(6), 3106鈥?117 (2013) CrossRef
  27. ViSOR datasets. http://www.openvisor.org
  28. H. Wang, D. Suter, K. Schindler, C. Shen, Adaptive object tracking based on an effective appearance filter. IEEE Trans. Pattern Anal. Mach. Intell. 29(9), 1661鈥?667 (2007) CrossRef
  29. X. Zhang, W. Hu, S. Maybank, X. Li, M. Zhu, Sequential particle swarm optimization for visual tracking, in / Proc. Int. Conf. Comp. Vis. Pattern Recognition (2008), pp. 1鈥?
  30. X. Zhang, W. Hu, W. Qu, S. Maybank, Multi-object tracking via species-based particle swarm optimization. IEEE Trans. Circuits Syst. Video Technol. 20(11), 1590鈥?602 (2010) CrossRef
  31. Y. Zhang, T. Hesketh, H. Wang et al., Actuator fault compensation for nonlinear systems using adaptive tracking control. Circuits Syst. Signal Process. 29(3), 419鈥?30 (2010) CrossRef
  32. H. Zulfiqar, Y.H. Gu, Joint feature correspondences and appearance similarity for robust visual object tracking. IEEE Trans. Inf. Forensics Secur. 5(3), 591鈥?06 (2010) CrossRef
  
• 作者单位：Xu Cheng (1)
  Nijun Li (1)
  Suofei Zhang (1)
  Zhenyang Wu (1)
  
  1. School of Information Science and Engineering, Southeast University, Nanjing, 210096, China
  
• ISSN：1531-5878

文摘

We propose a novel approach for visual tracking based on a particle swarm optimization (PSO) framework using SIFT feature points correspondence and multiple fragments in a candidate target region to cope with the problems of partial occlusions, illumination changes, and large motion changes of the tracked target. Firstly, optimal search in the successive frame tracking process is performed by the PSO algorithm, which guides all particles towards the global optima state based on a fitness function. Then, the SIFT feature information is integrated into the iterative results of PSO to acquire a more accurate tracking state. Secondly, we present an effective appearance model updating criterion, which evaluates which fragments in appearance model need updating at each frame. However, the fragments with occluded parts or low quality measure values are not updated. The method for updating appearance model is introduced to improve the tracking performance. Compared with state-of-the-art algorithms, the proposed method can still stably track the target during the course of long-term partial occlusions using superior fragments of tracked target. The experiment results demonstrate the effectiveness of our algorithm in complex environments where the target object undergoes partial occlusions and large changes in pose and illumination.