检测识别跟踪分离的在线多样本视频目标跟踪

doi:10.3969/j.issn.1000-3428.2012.23.048

计算机工程 ›› 2012, Vol. 38 ›› Issue (23): 194-197. doi: 10.3969/j.issn.1000-3428.2012.23.048

检测识别跟踪分离的在线多样本视频目标跟踪

刘国营，陈秀宏，庄甘霖

(江南大学数字媒体学院，江苏无锡 214122)

收稿日期:2012-02-07 出版日期:2012-12-05 发布日期:2012-12-03
作者简介:刘国营(1986－)，男，硕士研究生，主研方向：机器视觉，数字图像处理；陈秀宏，教授、博士；庄甘霖，硕士研究生
基金资助:
中央高校基本科研业务费专项基金资助项目(JUSRP211A70)

Online Multiple Sample Video Target Tracking with Separation of Detection Recognition and Tracking

LIU Guo-ying, CHEN Xiu-hong, ZHUANG Gan-lin

(School of Digital Media, Jiangnan University, Wuxi 214122, China)

Received:2012-02-07 Online:2012-12-05 Published:2012-12-03

摘要/Abstract

摘要： 在视频目标跟踪中，经常出现“漂移”现象，且学习算法需要离线训练。为此，提出一种检测识别跟踪分离的在线多样本视频目标跟踪方法。利用多样本学习方法解决学习算法在更新过程中的内在不确定性，使用优于在线半监督的boosting方法解决“漂移”现象。实验结果表明，该方法鲁棒性较好，可以有效解决“漂移”现象，并能实时地完成在线跟踪。

关键词: 多样本, 分类器, 在线学习, 目标跟踪, boosting方法

Abstract: There is the drift phenomenon in video target tracking, and learning algorithm needs offline training. In order to solve this problem, a online multiple sample video target tracking method for separation of detection, recognition and tracking is proposed. In this method, multiple sample approach tries to resolve the inherent ambiguities of many practical learning in the process of update instance. And the beyond semi-supervised learning tends to settle the “drifting” problem. Experimental results show that this method is more robust, can effectively solve the “drifting” problem, and track the target real-time online.

Key words: multiple sample, classifier, online learning, target tacking, boosting method

中图分类号:

TP399

刘国营, 陈秀宏, 庄甘霖. 检测识别跟踪分离的在线多样本视频目标跟踪[J]. 计算机工程, 2012, 38(23): 194-197.

LIU Guo-Ying, CHEN Xiu-Hong, PENG Gan-Lin. Online Multiple Sample Video Target Tracking with Separation of Detection Recognition and Tracking[J]. Computer Engineering, 2012, 38(23): 194-197.

http://www.ecice06.com/CN/Y2012/V38/I23/194

参考文献

[1] 袁宝峰, 吴乐华, 曾伟, 等. 一种对光照具有鲁棒性的人脸跟踪算法[J]. 信号处理, 2012, 26(1): 132-136.
[2] Isard M, Maccormick J. Bramble: A Bayesian Multiple-blob Tracker[C]//Proc. of International Conference on Computer Vision. Vancouver, Canada: [s. n.], 2001.
[3] Lepetit V, Fua P. Keypoint Recognition Using Randomized Trees[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(9): 1465-1479.
[4] Grabner H, Bischof H. On-line Boosting and Vision[C]//Proc. of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA: [s. n.], 2006.
[5] Matthews I, Ishikawa T, Baker S. The Template Update Pro- blem[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(6): 810-815.
[6] Grabner M, Grabner H, Bischof H. Learning Features for Tracking[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. Minneapolis, USA: [s. n.], 2007.
[7] Grabner H, Leistner C, Bischof H. Semi-supervised On-line Boosting for Robust Tracking[C]//Proc. of European Conference on Computer Vision. Marseille, France: [s. n.], 2008.
[8] Stalder S, Grabner H, Gool L V. Beyond Semi-supervised Tracking: Tracking Should be as Simple as Detection, but not Simpler than Recognition[C]//Proc. of International Conference on Computer Vision Workshops. Kyoto, Japan: [s. n.], 2009.
[9] Viola P, Platt J C, Zhang Cha. Multiple Instance Boosting for Object Detection[EB/OL]. (2010-11-21). http://research.microsoft. com/apps/pubs/?id=69348.
[10] Viola P, Jones M. Rapid Object Detection Using a Boosted Cascade of Simple Features[C]//Proc. of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE Press, 2001.
[11] Oza N C. Online Ensemble Learning[D]. Berkeley, USA: University of California, 2001.
[12] Grabner H, Grabner M, Bischof H. Real-time Tracking via Online Boosting[C]//Proc. of British Machine Vision Conference. Edinburgh, UK: [s. n.], 2006.
[13] Bernhard Z, Christian L, Amir S, et al. On-line Semi-supervised Multiple-instance Boosting[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. San Francisco, USA: [s. n.], 2010.
[14] Babenko B, Yang Ming-Hsuan, Belongie S. Visual Tracking with Online Multiple Instance Learning[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. Miami, USA: [s. n.], 2009.
[15] Lin Zhe, Hua Gang, Davis L. Multiple Instance Feature for Robust Part-based Object Detection[C]//Proc. of IEEE Conference on Computer Vision and Pattern Recognition. Miami, USA: [s. n.], 2009.

选择文件类型/文献管理软件名称

选择包含的内容

检测识别跟踪分离的在线多样本视频目标跟踪

Online Multiple Sample Video Target Tracking with Separation of Detection Recognition and Tracking

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	谢虹, 姜文刚. RRA-InceptionV3结合鲁棒稀疏表示的表情识别方法[J]. 计算机工程, 2023, 49(7): 196-203.
[2]	王春雷, 张建林, 李美惠, 徐智勇, 魏宇星. 结合卷积Transformer的目标跟踪算法[J]. 计算机工程, 2023, 49(4): 281-288,296.
[3]	周海赟, 项学智, 王馨遥, 任文凯. 多特征融合的端到端链式行人多目标跟踪网络[J]. 计算机工程, 2022, 48(9): 305-313.
[4]	鲁统伟, 徐子昕, 闵锋. 基于生成对抗网络的知识蒸馏数据增强[J]. 计算机工程, 2022, 48(4): 70-80.
[5]	李莉, 任振康, 石可欣. 代价敏感的Boosting软件缺陷预测方法[J]. 计算机工程, 2022, 48(3): 175-180.
[6]	杨帅东, 谌海云, 徐钒诚, 赵书朵, 袁杰敏. 基于孪生区域建议网络的无人机目标跟踪算法[J]. 计算机工程, 2022, 48(1): 288-295,304.
[7]	钱声攀, 于洋, 翟天一, 张徐东, 常彦博. 基于深度学习的主机负载在线预测模型研究[J]. 计算机工程, 2021, 47(9): 84-89.
[8]	张海涛, 秦鹏程. 基于GMS与FPME的视频目标跟踪方法[J]. 计算机工程, 2021, 47(7): 226-231.
[9]	任立成, 杨嘉棋, 魏宇星, 张建林. 基于特征融合与双模板嵌套更新的孪生网络跟踪算法[J]. 计算机工程, 2021, 47(7): 239-248.
[10]	苏超群, 朱正为, 郭玉英. 基于高效卷积算子的异常抑制目标跟踪算法[J]. 计算机工程, 2021, 47(7): 266-272,288.
[11]	李志鹏, 张睿. 一种基于对抗学习的实时跟踪模型设计[J]. 计算机工程, 2021, 47(6): 262-270.
[12]	尚桠朝, 孟令军. 基于多特征和尺度估计的KCF_MTSA算法[J]. 计算机工程, 2021, 47(3): 102-108,116.
[13]	罗朔, 侯进, 谭光鸿, 韩雁鹏. 一种低参数的孪生卷积网络实时目标跟踪算法[J]. 计算机工程, 2021, 47(2): 84-89.
[14]	刘菲菲, 伍忠东, 丁龙斌, 张凯. 基于改进在线序列极限学习机的AMI入侵检测算法[J]. 计算机工程, 2020, 46(9): 136-142,148.
[15]	张腾飞, 周书仁, 彭建. 基于双孪生网络的自适应选择跟踪系统[J]. 计算机工程, 2020, 46(6): 103-107.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

检测识别跟踪分离的在线多样本视频目标跟踪

Online Multiple Sample Video Target Tracking with Separation of Detection Recognition and Tracking

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价