文章快速检索 高级检索

Object tracking based on the joint model using L2-norm minimization
WANG Meng, WU Yi, DENG Jiankang, LIU Qingshan
School of Information & Control, Nanjing University of Information Science and Technology, Nanjing 210044, China
Abstract:The computational cost of the tracking algorithm based on the sparse representation is so much large, at the same time, the target apparence changes on account of a variety of reasons,which makes the object tracking process complicated and time consuming. A joint model is reasonably proposed by combining the global template based on the discriminant model and the generation model based on the local descriptor, properly solved by the L2-norm minimization solution in a bayesian inference framework, which is proved to be effective and efficient. In the process of the object tracking process, the plus template and the minus template of the discriminant model and the coefficient vector of the generative model are timely updated so as to have a strong adaptability and robust discrimination. The experimental results finally show that compared with other typical algorithms, the proposed algorithm has stronger robustness in the case of illumination, scale changes, shelter, rotation and so on.
Key words: object tracking     L2-norm minimization     discriminative model     generative model     subspace

L2范数约束项的作用有2个:①它使解b具有一定的稀疏度,但是L2范数的稀疏度远低于L1范数的稀疏度.②它使得最小化的解更加稳定.L2范数最小化很容易求解,令||UTby||22+λ||b||22的导数为0,即,可得出:

P=(UTU+λI)－1UT,很显然,P是独立于y的,所以对于通过粒子滤波得到的候选样本只需要计算一次P.如果把所有的候选样本看作一个向量集Y,则所有候选样本的表观系数可以一次性求得:

 图 1 利用分块得到的图像块Fig. 1 Image blocks by the sliding window

2.1 目标运动模型

2.3 模型更新策略

 图 2 总体流程图Fig. 2 Overall flow chart
3 实验结果与分析

 图 3 不同算法在不同测试序列上的跟踪结果Fig. 3 Tracking results with different algorithms on different videos

 图 4 针对不同的测试视频的跟踪误差曲线结果对比Fig. 4 Tracking error curve result according to different test videos
 图 5 针对不同的测试视频的重叠率曲线结果对比Fig. 5 Overlap rate curve result according to different test videos
4 结 论

 [1] Wu Y, Lim J,Yang M H.Online object tracking:a benchmark[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Washington,DC:IEEE Computer Society,2013:2411-2418. Click to display the text [2] 邵文坤,黄爱民, 韦庆.目标跟踪方法综述[J].影像技术,2006(1):17-20. Shao W K,Huang A M,Wei Q.Target tracking method review[J].Image Technology,2006(1):17-20(in Chinese). Cited By in Cnki (121) [3] Zhong W, Lu H,Yang M H.Robust object tracking via sparsity-based collaborative model[C]//Proc IEEE Comput Soc Conf Comput Vision Pattern Recognition.Washington,DC:IEEE Computer Society,2012:1838-1845. Click to display the text [4] 沈丁成,薛彦兵, 张桦,等.一种鲁棒的基于在线boosting目标跟踪算法研究[J].光电子·激光,2013,24(11):30. Shen D C,Xue Y B,Zhang H,et al.A robust online boosting target tracking algorithm based on the research[J].Journal of Photoelectron·Laser,2013,24(11):30(in Chinese). Cited By in Cnki (4) [5] Grabner H, Grabner M,Bischof H.Real-time tracking via on-line boosting[C]//BMVC 2006-Proceedings of the British Machine Vision Conference 2006.Edinburgh:British Machine Vision Association,2006:47-56. Click to display the text [6] 张颖颖,王红娟, 黄义定.基于在线多实例学习的跟踪研究[J].南阳师范学院学报,2012,10(12):35-37. Zhang Y Y,Wang H J,Huang Y D.Based on multiple instance learning online tracking study[J].Journal of Nanyang Normal University,2012,10(12):35-37(in Chinese). Cited By in Cnki (1) [7] Babenko B, Belongie S,Yang M H.Visual tracking with online multiple instance learning[C]//2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.Piscataway,NJ:IEEE Computer Society,2009:983-990. Click to display the text [8] Avidan S. Ensemble tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(2):261-271. Click to display the text [9] Ross D, Lim J,Lin R S,et al.Incremental learning for robust visual tracking[J].International Journal of Computer Vision,2008,77(1):125-141. Click to display the text [10] 齐飞,罗予频, 胡东成.基于均值漂移的视觉目标跟踪方法综述[J].计算机工程,2007,33(21):24-27. Qi F,Luo Y P,Hu D C.Visual target tracking method based on mean shift review[J].Computer Engineering,2007,33(21):24-27(in Chinese). Cited By in Cnki (43) [11] Black M,Jepson A. Eigentracking:robust maching and tracking of articulated objects using a view based representation[J].International Journal of Computer Vision,1998,26(1):63-84. Click to display the text [12] Yang A Y, Sastry S S,Ganesh A,et al.Fast 1 -minimization algorithms and an application in robust face recognition:a review[C]//Image Processing.Hong Kong:IEEE,2010:1849-1852. [13] Wright J, Yang A Y,Ganesh A,et al.Robust face recognition via sparse representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(2):210-227. Click to display the text [14] Mei X, Ling H.Robust visual tracking using L1 minimization[C]//Computer Vision.Anchorage,Alaska:IEEE,2009:1436-1443. Click to display the text [15] Mei X, Ling H,Wu Y,et al.Minimum error bounded efficientl1 tracker with occlusion detection[C]//Computer Vision and Pattern Recognition.Colorado Springs:IEEE,2011:1257-1264. Click to display the text [16] Bao C L, Wu Y,Ling H,et al.Real time robust l1 tracker using accelerated proximal gradient approach[C]//Proc IEEE Comput Soc Conf Comput Vision Pattern Recognition.Washington,DC:IEEE Computer Society,2012:1830-1837. Click to display the text [17] Zhang T Z, Ghanem B,Liu S,et al.Robust visual tracking via multi-task sparse learning[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Washington,DC:IEEE Computer Society,2012:2042-2049. Click to display the text [18] Zhang D, Yang M,Feng X.Sparse representation or collaborative representation:which helps face recognition?[C]//Computer Vision,2011:471-478. Click to display the text [19] Xiao Z Y, Lu H,Wang D.Object tracking with L2-RLS[C]//Proceedings-International Conference on Pattern Recognition.Piscataway,NJ:Institute of Electrical and Electronics Engineers Inc,2012:1351-1354. Click to display the text [20] Adam A, Rivlin E,Shimshoni I.Robust fragments-based tracking using the integral histogram[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.New York:Electronics Engineers Computer Society,2006:798-805. Click to display the text [21] Kwon J, Lee K M.Visual tracking decomposition[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Computer Society,2010:1269-1276. Click to display the text [22] Maggio E, Cavallaro A.Hybrid particle filter and mean shift tracker with adaptive transition model[C]//ICASSP,IEEE International Conference on Acoustics,Speech and Signal Processing-Proceedings.Philadelphia,PA:Institute of Electrical and Electronics Engineers Inc,2005:221-224. Click to display the text

#### 文章信息

WANG Meng, WU Yi, DENG Jiankang, LIU Qingshan

Object tracking based on the joint model using L2-norm minimization

Journal of Beijing University of Aeronautics and Astronsutics, 2015, 41(3): 559-566.
http://dx.doi.org/10.13700/j.bh.1001-5965.2014.0455