收稿日期: 2015-08-29
修回日期: 2015-12-13
网络出版日期: 2016-07-30
基金资助
国家自然科学基金(No.61203273)；江苏省自然科学基金(No.BK20141004)；江苏省普通高校自然科学研究基金(No.11KJB510009，No.14KJB510019)；江苏省“信息与通信工程”优势学科项目；江苏省大学生实践创新训练计划项目基金（No.201510300036Z）资助
Target Tracking Based on Online Template Clustering
Received date: 2015-08-29
Revised date: 2015-12-13
Online published: 2016-07-30
针对基于模板的目标跟踪算法存在模板冗余高、难以适应非刚性目标外观多变的问题，提出一种基于粒子滤波的模板在线聚类目标跟踪方法.首先建立用于描述目标和背景的正、负模板集，然后抽取候选粒子，使用候选粒子与正、负模板集的类内距离以及正、负模板集之间的类间距离来构建似然函数，最后依据最大后验概率准则确定最佳候选粒子作为跟踪结果.根据视频序列中连续变化的目标状态，将一定范围内的相似目标状态视为一个状态类，确定当前状态类的聚类半径.采用均值漂移算法对正模板集及最近几帧跟踪结果进行聚类，并将聚类后的中心集作为新的正模板集.实验表明，该算法能保留目标不同的外观状态，在复杂情况下仍能准确跟踪目标.
胡昭华, 王冠南, 王珏, 邵晓雯, 卞飞飞 . 模板在线聚类的目标跟踪[J]. 应用科学学报, 2016 , 34(4) : 430 -440 . DOI: 10.3969/j.issn.0255-8297.2016.04.008
The template of most tracking algorithms is difficult to suit appearance changes of non-rigid objects due to high redundancy. To deal with the problem, a tracking method with online templates clustering based on particle filter is proposed. Positive and negative template sets are established to describe the target and background. Candidates are then extracted based on the dynamic model. A likelihood function is built based on the withinclass distance between the candidate and template sets, and the between-class distance between positive sets and negative sets. The best candidate is considered the tracking result according to the maximum a posteriori probability (MAP). The main idea of online template clustering is as follows: First, the cluster radius is determined by the state class produced from a series of continuous target states in a certain range. Second, the positive template set combined with the recent several tracking results are used for clustering by using the mean shift iterative method and the above cluster radius. Third, the updated positive template set consists of the new cluster centers. In complicated situations, experiments show that this method can retain different appearance states of target and track the target accurately.
[1] Zhong W, Lu H C, Yang M. Robust object tracking via sparse collaborative appearance model[J]. IEEE Transactions on Image Processing, 2014, 23(5): 2356-2368.
[2] 贾旭. 基于局部外观模型的目标跟踪方法研究[D]. 大连:大连理工大学, 2013.
[3] Zhang K H, Zhang L, Yang M. Real-time compressive tracking [C]//Proceeding of the European Conference on Computer Vision, Firenze, 2012: 864-877.
[4] Henriques J F, Caseiro R, Martins P. Exploiting the circulant structure of tracking-bydetection with kernels[M] . Berlin, Heidelberg, Germany: Springer Press, 2012: 702-715.
[5] Danelljan M, Khan F S, Felsberg M. Adaptive color attributes for real-time visual tracking[C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 2014: 1090-1097.
[6] 侯跃恩. 基于稀疏表示的视觉目标跟踪算法研究[D]. 广州:华南理工大学, 2014.
[7] 张旭东, 陈仲海, 胡良梅. 基于联合模板稀疏表示的目标跟踪方法[J]. 控制与决策, 2015, 30(9): 1696-1700. Zhang X D, Chen Z H, Hu L M. Target tracking joint template based on sparse representation[J]. Control and Decision, 2015, 30 (9): 1696-1700. (in Chinese)
[8] 杨晓玲. 基于粒子滤波和稀疏表示的视频目标跟踪[J]. 信息技术, 2015, 39(6): 103-108. Yang X L. Target tracking based on particle filter joint sparse representation [J]. Information Technology, 2015, 39(6): 103-108. (in Chinese)
[9] Ross D A, Lim J, Lin R S. Incremental learning for robust visual tracking [J]. International Journal of Computer Vision, 2008, 77(1/3): 125-141.
[10] Xiao Z Y, Lu H C, Wang D. L2-RLS-based object tracking [J]. IEEE Transactions on Circuits and Systems for Video Technology, 2014, 24(8): 1301-1309.
[11] Zhuang B H, Lu H C, Xiao Z Y. Visual tracking via discriminative sparse similarity map [J]. IEEE Transactions on Image Processing, 2014, 23(4): 1872-1881.
[12] Bao C L, Wu Y, Ling H B. Real time robust l1 tracker using accelerated proximal gradient approach [C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 1830-1837.
[13] Firouzi H, Najjaran H. Adaptive on-line similarity measure for direct visual tracking [J]. IEEE Image and Vision Computing, 2014, 32(4): 227-236.
[14] 汤贝. 基于模板匹配的目标跟踪算法研究[D]. 长沙:长沙理工大学, 2013.
[15] Fukunaga K, Hostetler L D. The estimation of the gradient of a density function with applications in pattern recognition [J]. IEEE Transactions on Information Theory, 1975, 21(1): 32-40.
[16] Wu Y, Lim J, Yang M. Online object tracking: a benchmark [C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Portland, 2013: 2411-2418.
[17] Zhong W, Lu H C, Yang M. Robust object tracking via sparsity-based collaborative model[C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 1838-1845.
[18] Hare S, Saffari A, Torr P H S. Structured output tracking with kernels [C]//Proceeding of IEEE International Conference on Computer Vision, Barcelona, 2011: 263-270.
[19] Zhang T Z, Ghanem B, Liu S. Robust visual tracking via multi-task sparse learning[C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 2042-2049.
[20] Oron S, Bar H A, Levi D. Locally orderless tracking [C]// Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 1940-1947.
[21] Jia X, Lu H C, Yang M. Visual tracking via adaptive structural local sparse appearance model[C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 1822-1829.
[22] Dinh, Thang B, Vo N. Context tracker: exploring supporters and distracters in unconstrained environments [C]//Proceeding of IEEE Conference on Computer Vision and Pattern Recognition, Colorado Springs, 2011: 1177-1184.
/
| 〈 |
|
〉 |
