基于稀疏表示和特征选择的LK目标跟踪

被引:5
作者
潘晴
曾仲杰
机构
[1] 广东工业大学信息工程学院
基金
广东省自然科学基金;
关键词
视觉跟踪; 稀疏表示; LK图像配准算法; 特征选择;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
为了实现复杂场景中的视觉跟踪,提出了一种以LK(Lucas-Kanade)图像配准算法为框架,基于稀疏表示的在线特征选择机制。在视频序列的每一帧,筛选出一些能够很好区分目标及其相邻背景的特征,从而降低干扰对跟踪的影响。该算法分别构造前景字典和背景字典,前景字典来自于第一帧的手动标定,并随着跟踪结果不断更新,而背景字典则在每一帧重新构造。同时,一种新的字典更新策略不仅能有效应对目标的外观变化,而且通过特征选择机制,能避免在更新过程中引入干扰,从而克服了漂移现象。大量的实验结果表明,该算法能有效应对视角变化、光照变化以及大面积的局部遮挡等挑战。
引用
收藏
页码:625 / 628
页数:4
相关论文
共 14 条
[1]  
Locality-constrained linear coding for image classification. Wang Jinjun,Yang Jianchao,Yu Kai,et al. Proceedings of the 23rd IEEE Conference on Computer Vision and Pattern Recognition (CVPR’’10) . 2010
[2]  
Robust visual tracking using L1 minimization. MEI Xue,LING Hai-bing. Proc of the 12th IEEE International Conference on Computer Vision . 2009
[3]  
Lucas-Kanade 20 Years On: A Unifying Framework[J] . Simon Baker,Iain Matthews. &nbspInternational Journal of Computer Vision . 2004 (3)
[4]  
Incremental Learning for Robust Visual Tracking[J] . David A. Ross,Jongwoo Lim,Ruei-Sung Lin,Ming-Hsuan Yang. &nbspInternational Journal of Computer Vision . 2008 (1)
[5]  
Object tracking[J] . Alper Yilmaz,Omar Javed,Mubarak Shah. &nbspACM Computing Surveys (CSUR) . 2006 (4)
[6]  
Beyond semi-supervised tracking:tracking should be as simple as detection,but not simpler than recognition. STALDER S,GRABNER H,Van GOOL L. Proc of the 12th IEEE International Conference on Computer Vision . 2009
[7]  
Real-time tracking via on-line boosting. GRABNER H,GRABNER M,BISCHOF H. Proc of British Machine Vision Conference . 2006
[8]  
Nonlocally Centralized Sparse Representation for Image Restoration. Dong, W.,Zhang, L.,Shi, G.,LI, X. Image Processing, IEEE Transactions on . 2013
[9]  
Robust Infrared Vehicle Tracking across Target Pose Change using L1Regularization. Ling H,Bai L,Blasch E, etal. Int. Conf. on Info Fusion . 2010
[10]  
Robust Visual Tracking and Vehicle Classification via Sparse Representation. Xue Mei,Haibin Ling. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2011