在线学习多重检测的可视对象跟踪方法

doi:10.3969/j.issn.0372-2112.2014.05.007

电子学报

在线学习多重检测的可视对象跟踪方法

权伟¹, 陈锦雄¹, 余南阳²

1. 西南交通大学信息科学与技术学院,四川成都 610031;
2. 西南交通大学机械工程学院,四川成都 610031

收稿日期:2013-03-28 修回日期:2013-07-12 出版日期:2014-05-25
- 作者简介:
- 权伟(通信作者) 男,1982年出生于四川宜宾,博士研究生.目前主要从事计算机视觉、模式识别、机器学习以及图像处理方面的研究. E-mail:xweiquan@gmail.com陈锦雄男,1962年出生于四川成都,博士.美国乔治梅森大学教授,主要从事计算机图形、虚拟现实、计算机视觉以及数据可视化和仿真等方面的研究.
- 基金资助:
- 国家自然科学基金 (No.40672203); 西南交通大学博士创新基金

Online Learning of Multiple Detectors for Visual Object Tracking

QUAN Wei¹, CHEN Jin-xiong¹, YU Nan-yang²

1. School of Information Science and Technology, Southwest Jiaotong University, Chengdu, Sichuan 610031, China;
2. School of Mechanical Engineering, Southwest Jiaotong University, Chengdu, Sichuan 610031, China

Received:2013-03-28 Revised:2013-07-12 Online:2014-05-25 Published:2014-05-25
- Supported by:
- National Natural Science Foundation of China (No.40672203); Fund for Doctoral Innovation of Southwest Jiaotong University

摘要/Abstract

摘要： 为了研究无约束环境下长时间可视跟踪问题,提出了一种在线学习多重检测的对象跟踪方法.该方法以随机蕨作为基础检测器结构,通过在线学习的方式,将目标对象的整体和局部表观,以及由场景学习中发掘的同步对象同时作为检测学习的基础数据,该检测器因而具备了对这多种对象的独立检测能力.由于其各个检测部分发挥了各自不同的作用,本文从测量的角度将检测器对这三种对象检测的结果进行融合,通过计算检测关于目标的配置概率进而确定目标位置,实现对象跟踪任务.基于真实视频序列的实验结果验证了本文方法的有效性和稳定性,以及较现有的跟踪方法在跟踪性能上的提高.

关键词: 对象跟踪, 多重检测, 在线学习, 随机蕨

Abstract: In order to study the problem of long-term visual tracking in unconstrained environments,this paper proposes a method of learning multiple detectors online for visual object tracking.The method uses the random ferns as the basic detector.The entire and the local appearances of the target and the connected objects which are explored by the context learning are used synchronously as the training data to build and upgrade the object detector on the fly.Thus it is able to detect the objects with different classes independently.Since different detections are related to different object classes,the results of object detections are fused as the measurements and the probabilities of configuration hypotheses for the measurements to the target are calculated to find the target location for visual tracking task.Experimental results based on the real-world video sequences validate the effectiveness and robustness of our approach and demonstrate its better tracking performance than several state-of-the-art methods.

Key words: object tracking, multiple detectors, online learning, random ferns

中图分类号:

TP319

权伟, 陈锦雄, 余南阳. 在线学习多重检测的可视对象跟踪方法[J]. 电子学报, DOI: 10.3969/j.issn.0372-2112.2014.05.007.

QUAN Wei, CHEN Jin-xiong, YU Nan-yang . Online Learning of Multiple Detectors for Visual Object Tracking[J]. Acta Electronica Sinica, DOI: 10.3969/j.issn.0372-2112.2014.05.007.

参考文献

[1] B Lucas,T Kanade.An iterative image registration technique with an application to stereo vision[A].Proceedings of the 7th International Joint Conference on Artificial Intelligence (IJCAI)[C].Vancouver,Canada,1981.674-679.
[2] M Isard,A Blake.CONDENSATION—Conditional density propagation for visual tracking[J].International Journal of Computer Vision (IJCV),1998,29(1):5-28.
[3] 常发亮,马丽,刘增晓,乔谊正.复杂环境下基于自适应粒子滤波器的目标跟踪[J].电子学报,2006,34(12):2150-2153. CHANG Fa-liang,MA Li,LIU Zeng-xiao,QIAO Yi-zheng.Target tracking based on adaptive particle filter under complex background[J].Acta Electronica Sinica,2006,34(12):2150-2153.(in Chinese)
[4] D Comaniciu,V Ramesh,P Meer.Kernel-based object tracking[J].IEEE Trans Pattern Analysis and Machine Intelligence,2003,25(5):564-5773.
[5] 江淑红,汪沁,张建秋,胡波.基于目标中心距离加权和图像特征识别的跟踪算法[J].电子学报,2006,34(7):1175-1180. JIANG Shu-hong,WANG Qin,ZHANG Jian-qiu,HU Bo.An image tracking algorithm based on object center distance-weighting and image feature recognition[J].Acta Electronica Sinica.2006,34(7):1175-1180.(in Chinese)
[6] 李培华.一种新颖的基于颜色信息的粒子滤波器跟踪算法[J].计算机学报,2009,32(12):2454-2463. Li Pei-Hua.A novel color based particle filter algorithm for object tracking[J].Chinese Journal of Computers,2009,32(12):2454-2463.(in Chinese)
[7] A Yilmaz,O Javed,M Shah.Object tracking:A survey[J].ACM Computing Surveys,2006,38(4):13.
[8] S Avidan.Ensemble tracking[J].IEEE Trans Pattern Analysis and Machine Intelligence,2007,29(2):261-271.
[9] R Collins,Y Liu,M Leordeanu.Online selection of discriminative tracking features[J].IEEE Trans Pattern Analysis and Machine Intelligence,2005,27(10):1631-1643.
[10] J Lim,D Ross,R Lin,M Yang.Incremental learning for visual tracking[A].Proceedings of Conference on Advances in Neural Information Processing Systems (NIPS)[C].Vancouver,Canada:the MIT Press,2004.793-800.
[11] Q Yu,T Dinh,G Medioni.Online tracking and reacquisition using co-trained generative and discriminative trackers[J].Lecture Notes in Computer Science,2008,5303:678-691.
[12] S Avidan.Support vector tracking[J].IEEE Trans Pattern Analysis and Machine Intelligence,2004,26(8):1064-1072.
[13] DA Ross,J Lim,R Lin,M Yang.Incremental learning for robust visual tracking[J].International Journal of Computer Vision (IJCV),2008,77(1-3):125-141.
[14] 温静,李洁,高新波.基于增量张量子空间学习的自适应目标跟踪[J].电子学报,2009,37(7):1618-1623. WEN Jing,LI Jie,GAO Xin-bo.Adaptive object tracking with incremental tensor subspace learning[J].Acta Electronica Sinica,2009,37(7):1618-1623.(in Chinese)
[15] H Grabner,H Bischof.On-line boosting and vision[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].New York,USA,2006,1:260-267.
[16] H Grabner,C Leistner,H Bischof.Semi-supervised on-line boosting for robust tracking[J].Lecture Notes in Computer Science,2008,5302:234-247.
[17] S Stalder,H Grabner,L van Gool,E Zurich,K Leuven.Beyond semi-supervised tracking:tracking should be as simple as detection,but not simpler than recognition[A].Proceedings of 12th IEEE International Conference on Computer Vision Workshops (ICCV Workshops)[C].Kyoto,Japan,2009.1409-1416.
[18] L Breiman.Random forests[J].Machine Learning,2001,45(1):5-32.
[19] C Leistner,A Saffari,J Santner,H Bischof.Semi-supervised random forests[A].Proceedings of 12th IEEE International Conference on Computer Vision (ICCV)[C].Kyoto,Japan,2009.506-513.
[20] C Leistner,A Saffari,H Bischof.MILForests:Multiple-Instance learning with randomized trees[A].Proceedings of 11th European Conference on Computer Vision (ECCV)[C].Crete,Greece,2010.29-42.
[21] A Saffari,C Leistner,J Santner,M Godec,H Bischof.On-line random forests[A].Proceedings of 12th IEEE International Conference on Computer Vision Workshops (ICCV Workshops)[C].Kyoto,Japan,2009.1393-1400.
[22] C Leistner,M Godec,A Saffari,H Bischof.On-line multi-view forests for tracking[J].Lecture Notes in Computer Science,2010,6376:493-502.
[23] A Wang,G Wan,Z Cheng,S Li.An incremental extremely random forest classifier for online learning and tracking[A].Proceedings of 16th IEEE International Conference on Image Processing (ICIP)[C].Cairo,2009.1449-1452.
[24] 王爱平,万国伟,程志全,李思昆.支持在线学习的增量式极端随机森林分类器[J].软件学报,2011,22(9):2059-2074. WANG Ai-Ping,WAN Guo-Wei,CHENG Zhi-Quan,LI Si-Kun.Incremental learning extremely random forest classifier for online learning[J].Journal of Software,2011,22(9):2059-2074.(in Chinese)
[25] B Babenko,M-H.Yang,S Belongie.Visual tracking with online multiple instance learning[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].Miami,2009.983-990.
[26] Z Kalal,J Matas,K Mikolajczyk.Online learning of robust object detectors during unstable tracking[A].Proceedings of 12th IEEE International Conference on Computer Vision Workshops (ICCV Workshops)[C].Kyoto,Japan,2009.1417-1424.
[27] Z Kalal,J Matas,K Mikolajczyk.P-N learning:bootstrapping binary classifiers by structural constraints[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].San Francisco,2010.49-56.
[28] M Ozuysal,P Fua,V Lepetit.Fast keypoint recognition in ten lines of code[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].Minneapolis,2007.1-8.
[29] J Gall,V Lempitsky.Class-specific Hough forests for object detection[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].Miami,2009.1022-1029.
[30] J Gall,N Razavi,L Van Gool.On-line adaption of class-specific codebooks for instance tracking[A].Proceedings of British Machine Vision Conference (BMVC)[C].Aberystwyth,2010.1-12.
[31] A Yao,J Gall,L Van Gool.A Hough transform-based voting framework for action recognition[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].San Francisco,2010.2061-2068.
[32] J Gall,A Yao,N Razavi,L van Gool,Victor Lempitsky.Hough forests for object detection,tracking,and action recognition[J].IEEE Trans Pattern Analysis and Machine Intelligence,2011,33(11):2188-2202.
[33] D H Ballard.Generalizing the Hough transform to detect arbitrary shapes[J].Pattern Recognition,1981,13(2):111-122.
[34] M Yang,Y Wu,G Hua.Context-aware visual tracking[J].IEEE Trans.Pattern Analysis and Machine Intelligence,2009,31(7):1195-1209.
[35] H Grabner,J Matas,L V Gool,P Cattin.Tracking the invisible:Learning where the object might be[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].San Francisco,2010.1285-1292.
[36] T B Dinh,N Vo,G Medioni.Context tracker:Exploring supporters and distracters in unconstrained environments[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].Providence,2011.1177-1184.
[37] J L Fan,Y Wu,S Y Dai.Discriminative spatial attention for robust tracking[A].Proceedings of 11th European Conference on Computer Vision (ECCV)[C].Crete,Greece,2010.480-493.
[38] M Godec,S Sternig,P M Roth,H Bischof.Context-driven clustering by multi-class classification in an active learning framework[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops)[C].San Francisco,2010.19-24.
[39] B Stenger,T Woodley,R Cipolla.Learning to track with multiple observers[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].Miami,2009.1063-6919.
[40] S Agarwal,A Awan,D Roth.Learning to detect objects in images via a sparse,part-based representation[J].IEEE Trans on Pattern Analysis and Machine Intelligence,2004,26(11):1475-1490.
[41] A Adam,E Rivlin,I Shimshoni.Robust fragments-based tracking using the integral histogram[A].Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR)[C].New York,USA,2006.798-805.

在线学习多重检测的可视对象跟踪方法

Online Learning of Multiple Detectors for Visual Object Tracking

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

[1]	陆克中, 陈超凡, 蔡桓, 吴定明. 面向概念漂移和类不平衡数据流的在线分类算法[J]. 电子学报, 2022, 50(3): 585-597.
[2]	罗大鹏, 杜国庆, 曾志鹏, 魏龙生, 高常鑫, 陈应, 肖菲, 罗琛. 基于少量样本学习的多目标检测跟踪方法[J]. 电子学报, 2021, 49(1): 183-191.
[3]	魏波, 吴瑞峰, 张文生, 吕敬钦, 王莹莹, 夏学文. 具有双重稀疏机制的在线学习算法[J]. 电子学报, 2019, 47(10): 2202-2210.
[4]	朱敬华, 尹旭明, 柏敬思, 王亚珂, 李金宝, 王雪. 传感器网络基于轨迹聚类的多目标跟踪算法[J]. 电子学报, 2017, 45(11): 2671-2676.
[5]	罗大鹏, 罗琛, 魏龙生, 韩家宝, 王勇, 马丽. 基于在线随机蕨分类器的实时视觉感知系统[J]. 电子学报, 2016, 44(5): 1139-1148.
[6]	李康, 何发智, 潘一腾, 孙航. 基于簇相似的多分类器目标跟踪算法[J]. 电子学报, 2016, 44(4): 821-825.
[7]	宋立锋;韦岗;王群生. 基于模板匹配的视频对象分割[J]. 电子学报, 2002, 30(7): 1075-1078.