计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230700113-8.doi: 10.11896/jsjkx.230700113

• 图像处理&多媒体技术 • 上一篇    下一篇

基于DIoU损失与平滑约束的结构化SVM目标跟踪方法

孙子文1, 袁广林2, 李从利1, 秦晓燕2, 朱虹2   

  1. 1 中国人民解放军陆军炮兵防空兵学院兵器工程系 合肥 230031
    2 中国人民解放军陆军炮兵防空兵学院信息工程系 合肥 230031
  • 发布日期:2024-06-06
  • 通讯作者: 袁广林(yuangl_plus@126.com)
  • 作者简介:(2215738260@qq.com)
  • 基金资助:
    安徽省自然科学基金(2008085QF325)

Object Tracking of Structured SVM Based on DIoU Loss and Smoothness Constraints

SUN Ziwen1, YUAN Guanglin2, LI Congli1, QIN Xiaoyan2, ZHU Hong2   

  1. 1 Department of Weaponry Engineering of PLA Army Academy of Artillery and Air Defense,Hefei 230031,China
    2 Department of Information Engineering of PLA Army Academy of Artillery and Air Defense,Hefei 230031,China
  • Published:2024-06-06
  • About author:SUN Ziwen,born in 1996,Ph.D candidate.His main research interests include computer vision and object trac-king.
    YUAN Guanglin,born in 1976,Ph.D,associate professor.His main research interests include computer vision,object tracking and machine learning.
  • Supported by:
    Natural Science Foundation of Anhui Province,China(2008085QF325)

PDF (PC)

258

摘要： 基于结构化SVM的目标跟踪因其优良的性能而受到了广泛的关注,但是现有方法存在损失函数不精确和模型漂移问题。针对这两个问题,首先提出基于DIoU损失与平滑约束的结构化SVM模型。该模型采用DIoU函数作为损失函数,利用t时刻超平面法向量wt与t－1时刻超平面法向量wt－1差值的L2范数作为平滑约束。其次基于对偶坐标下降原理设计了该模型的求解算法。最后利用提出的基于DIoU损失与平滑约束的结构化SVM实现了一种多尺度目标跟踪方法。对所提出的目标跟踪方法在OTB100和VOT-ST2021数据集上进行了实验验证,实验结果表明:所提出的Scale-DCSSVM在OTB数据集上的跟踪成功率比DeepSRDCF高1.1个百分点,在VOT-ST2021上的EAO比E.T.Track高1.2个百分点。所提方法具有较优的性能。

关键词: 目标跟踪, 结构化SVM, DIoU损失, 平滑约束

Abstract: Object tracking based on structured support vector machine has been widely concerned because of its excellent performance.However,the existing methods have the problems of imprecise loss function and model drift.To solve these two pro-blems,firstly,a structured SVM model is proposed based on DIoU loss and smoothness constraints.Secondly,DIoU function and L2 norm of the difference between wt and wt－1 are used respectively as the loss functions and the smoothness constraints in the model.Thirdly,the algorithm for the proposed model is designed with the dual coordinate descent principle.Finally,a multi-scale object tracking method is implemented via the proposed structured SVM on the basis of DIoU loss and smoothness constraints.The proposed object tracking method is experimentally validated on the OTB100 and VOT-ST2021 datasets,and the experimental results show that the tracking success rate of the Scale-DCSSVM on the OTB100 is 1.1％ higher than the DeepSRDCF,and the EAO on VOT-ST2021 is 1.2％ higher than the E.T.Track.The proposed object tracking method has superior performance.

Key words: Object tracking, Structured SVM, DIoU loss, Smoothness constraints

中图分类号: 

  • TP391.41
[1]ROSS D A,LIM J,LINR S,et al.Incremental Learning for Robust Visual Tracking[J].International Journal of Computer Vision,2008,77(1/2/3):125-141.
[2]MEI X,LING H.Robust visual tracking using L1 minimization[C]//Proceedings of IEEE Conference on Computer Vision.Kyoto,Japan:IEEE Computer Society Press,2009:1436-1443.
[3]SUN J,XIANG W,TAN S K,et al.Improved Kernel Correlation Filter Tracking Algorithm[J].Computer Engineering and Applications,2018,54(9):178-182.
[4]BABENKO B,YANG M H,BELONGIE S.Robust object tra-cking with online multiple instance learning[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(8):1619-1632.
[5]KALAL Z,MIKOLAJCZYK K,MATAS J.Tracking-learning-detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(7):1409-1422.
[6]SAFFARI A,LEISTNER C,SANTNER J,et al.Online random forests[C]//Proceedings of the IEEE Conference on Computer Vision.Kyoto,Japan:IEEE Computer Society Press,2009:1393-1400.
[7]HARE S,SAffARI A,TORR P H S.Struck:structured output tracking with kernels[C]//Proceedings of IEEE International Conference on Computer Vision.Barcelona,Spain:IEEE Computer Society Press,2011:263-270.
[8]WANG N,YEUNG D Y.Learning a Deep Compact Image Representation for Visual Tracking[C]//Proceedings of Neural Information Processing Systems.Curran Associates:IEEE Computer Society Press,2013:809-817.
[9]WEI Y Q,YANG X J.Kernel Correlation Filtering for VisualTracking of Deep Features[J].Computer Engineering and Applications,2020,56(15):147-152.
[10]AVIDAN S.Support vector tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(8):1064-1072.
[11]HARE S,SAffARI A,TORR P H S.Struck:structured output tracking with kernels[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,38(10):2096-2109.
[12]YAO R,SHI Q,SHEN C,et al.Robust Tracking with Weighted Online Structured Learning[C]//Proceedings of the European conference on Computer Vision.Berlin Heidelberg:Springer,2012,Part III:158-172.
[13]YAO R,SHI Q,SHEN C,et al.Part-based Visual Tracking with Online Latent Structural Learning[C]//Proceedings of Computer Vision and Pattern Recognition.Portland,USA:IEEE Computer Society Press,2013:2363-2370.
[14]BAI Y,TANG M.Robust Tracking via Weakly SupervisedRanking SVM[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Providence,RI,USA:IEEE Computer Society Press,2012:1854-1861.
[15]ZHANG J,MA S,SCLAROFF S.Meem:Robust tracking viamultiple experts using entropy minimization[C]//Proceedings of European Conference on Computer Vision.Zurich,Switzerland:IEEE Computer Society Press,2014:188-203.
[16]HONG S,YOU T,KWAK S,et al.Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network[C]//Proceedings of International Conference on Machine Learning.Lille,France:IEEE Computer Society Press,2015:597-606.
[17]NING J F,YANG J M,JIANG S J,et al.Object tracking viadual linear structured SVM and explicit feature map[C]//Procee-dings of IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE Computer Society Press,2016:4266-4274.
[18]WANG M M,LIU Y,HUANG Z Y.Large Margin ObjectTracking with Circulant Feature Maps[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,USA,2017:4800-4808.
[19]JI Z J,FENG K,QIANY H.Part-based Visual Tracking via Structural Support Correlation Filter[J].arXiv:1805.09971,2018.
[20]ZUO W,WU X,LIN L,et al.Learning Support Correlation Filters for Visual Tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,41(5):1158-1172.
[21]WANG C Y,WANG S Q,ZHANG L J,et al.A Multi featureDSST Target Tracking Algorithm Based on SVM Fusion[J].Ordnance Industry Automation,2021,40(7):39-45,66.
[22]QIAN Y N,CHEN Y W,LI G.Passive sonar target association tracking method based on historical motion feature constraints and SVM spectrum classification[J].Journal of Electronics and Information Technology,2023,45(8):2991-3001.
[23]ZHEN Z H,WANG P,LIU W,et al.Distance-IoU Loss:Faster and Better Learning for Bounding Box Regression[C]//Procee-dings of the AAAI Innovative Applications of Artificial Intelligence Conference.New York,USA,2020:12993-13000.
[24]RAMANAN D.Dual coordinate solvers for large-scale structu-ral svms[J].arXiv:1312.1743v2,2014.
[25]WU Y,LIM J,YANG M H.Object tracking benchmark[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1834-1848.
[26]ZABIH R,WOODFILL J.Non-parametric Local Transforms for Computing Visual Correspondence[C]//Proceedings of the European Conference on Computer Vision,Stockholm,Sweden.IEEE Computer Society Press,1994:151-158.
[27]KRISTAN M,MATAS J,LEONARDISA,et al.The Ninth Vi-sual Object Tracking[C]//Proceedings of the{VOT2021} Challenge Results International Conference on Computer Vision Workshops.Montreal,BC,Canada,2021:2711-2738.
[28]HONG S,YOU T,KWAK S,et al.Online Tracking by Learning Discriminative Saliency Map with Convolutional Neural Network[C]//Proceedings of the International Conference on Machine Learning.Lille:IEEE Computer Society Press,2015:597-606.
[29]BERTINETTO L,VALMADRE J,HENRIQUES J F,et al.Fully-Convolutional Siamese Networks for Object Tracking[C]//Proceedings of the European Conference on Computer Vision.Amsterdam,The Netherlands:IEEE Computer Society Press,2016:850-865.
[30]SONG Y B,MA C,GONG L J,et al.Convolutional ResidualLearning for Visual Tracking[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice,Italy:IEEE Computer Society Press,2017:2574-2583.
[31]DANELLJAN M,HAGER G,KHAN F S,et al.Convolutional Features for Correlation Filter Based Visual Tracking[C]//Proceedings of the IEEE International Conference on Computer Vision.Santiago:IEEE Computer Society Press,2015:621-629.
[32]MA C,HUANG J B,YANG X K,et al.Hierarchical Convolutional Features for Visual Tracking[C]//Proceedings of the IEEE International Conference on Computer Vision.Santiago,Chile:IEEE Computer Society Press,2015:3074-3082.
[33]HONG Z B,CHEN Z,WANG C H,et al.Multi-Store Tracker:ACognitive Psychology Inspired Approach to Object Tracking[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.Boston,USA:IEEE Computer Society Press,2015:749-758.
[34]DANELLJAN M,GUSTAV H,KHAN F S,et al.LearningSpatially Regularized Correlation Filters for Visual Tracking[C]//Proceedings of the IEEE International Conference on Computer Vision.Santiago:IEEE Computer Society Press,2015:4310-4318.
[35]BERTINETTO L,VALMADRE J,GOLODETZ S,et al.Staple:Complementary Learners for Real-Time Tracking[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.Las Vegas,NV:IEEE Computer Society Press,2016:1401-1409.
[36]LI B,WU W,WANG Q,et al.SiamRPN++:Evolution of Siamese Visual Tracking With Very Deep Networks[C]//Procee-dings of the Conference on Computer Vision and Pattern Recognition,Long Beach,CA,USA.2019:4282-4291.
[37]LUKEZIC A,MATAS J,KRISTAN M.D3S-A discriminative single shot segmentation tracker[C]//Proceedings of the Conference on Computer Vision and Pattern Recognition.Seattle,WA,USA,2020:7131-7140.
[38]LI X,MA C,WU B Y,et al.Target-Aware Deep Tracking[C]//Proceedings of the International Conference on Computer Vision and Pattern Recognition.CA,USA,2019:1369-1378.
[39]YAN B,PENG H W,WU K,et al.Lighttrack:Finding lightweight neural networks for object tracking via one-shot architecture search[C]//Proceedings of the Conference on Computer Vision and Pattern Recognition.Virtual,2021:15180-15189.
[40]BLATTER P,KANAKIS M,DANELLJAN M,et al.Efficient Visual Tracking with Exemplar Transformers[C]//Procee-dings of the Winter Conference on Applications of Computer Vision,Waikoloa,HI,USA.2023:1571-1581.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发