计算机科学

计算机科学 ›› 2019, Vol. 46 ›› Issue (3): 97-102.doi: 10.11896/j.issn.1002-137X.2019.03.013

• 2018 中国多媒体大会 • 上一篇    下一篇

一种基于2D和3D联合信息的改进MDP跟踪算法

王正宁1,周阳1,吕侠1,曾凡伟1,张翔1,张锋军2   

  1. (电子科技大学信息与通信工程学院 成都 611731)1
    (中国电子科技集团公司第三十研究所 成都 610041)2
  • 收稿日期:2018-07-02 修回日期:2018-09-10 出版日期:2019-03-15 发布日期:2019-03-22
  • 作者简介:王正宁(1979－),男,博士,副教授,主要研究领域为图像及视频处理、智能交通系统、多媒体通讯系统及应用,E-mail:zhengning.wang@uestc.edu.cn;周阳(1991－),男,硕士生,主要研究领域为基于视觉的目标检测与跟踪算法;吕侠(1994－),男,硕士生,主要研究领域为基于视觉的目标检测与跟踪算法;曾凡伟(1994－),男,硕士生,主要研究领域为图像压缩与拼接;张翔(1995－),男,硕士生,主要研究领域为图像压缩与拼接;张锋军(1975－),男,研究员级高工,主要研究领域为软件工程和网络安全。
  • 基金资助:
    四川科技厅项目(2018GZ0071)资助

Improved MDP Tracking Method by Combining 2D and 3D Information

WANG Zheng-ning1,ZHOU Yang1,LV Xia1,ZENG Fan-wei1,ZHANG Xiang1,ZHANG Feng-jun2   

  1. (School of Information and Communication Engineering,University of Electronic Science and Technology of China,Chengdu 611731,China)1
    (The 30th Research Institute of CETC,Chengdu 610041,China)2
  • Received:2018-07-02 Revised:2018-09-10 Online:2019-03-15 Published:2019-03-22

PDF (PC)

780

摘要： 在线多目标跟踪算法是自动驾驶和辅助驾驶系统的重要组成部分。目前,大部分多目标跟踪方法集中于图像域跟踪。虽然通过建立自适应在线模型或最小化能量函数可以解决大多数跟踪问题,但是如何处理复杂交通场景下目标的相互遮挡仍是研究者们面临的难题。文中基于2D和3D联合信息提出了一种改进的基于马尔科夫决策过程(MDP)的跟踪算法,通过将原始MDP跟踪算法的相似性特征由图像域拓展到空间域,使用一种新的光流特征描述子即多图像前后向跟踪误差(Multi-image FB error)来代替原算法的多区域前后向跟踪误差(Multi-aspect FB error),取得了良好的跟踪效果。最后,采用KITTI数据库对本文算法进行测试,结果显示其综合性能相较于原算法有显著提升。

关键词: 2D-3D联合特征, 多目标跟踪, 多图像光流, 马尔科夫决策过程

Abstract: Online multi-object tracking (MOT) plays an important role in autonomous driving and ADAS system.Most of recent MOT methods concentrate on tracking in image domain.Although they can solve most of problems by building adaptive online models or optimizing energy functions,it’s still an obstacle for researchers to handle mutual occlusion in complex traffic scenes.In this paper,an improved tracking method was proposed by introducing 3D information to the Markov decision processes (MDP) tracker.The original MDP similarity feature was extended from image domain to spatial domain with 2D-3D combined feature,and a new optical flow descriptor,called multi-image FB error,was addressed to replace the original multi-aspect FB error.This methodwas tested on KITTI benchmark and the results verified that the comprehensive performance of the proposed method is refined significantly in comprehensive performance compared with the original method.

Key words: 2D-3D combined feature, Markov decision process, Multi-image optical flow, Multi-object tracking

中图分类号: 

  • TP181
[1]XIANG Y,CHOI W,LIN Y,et al.Subcategory-Aware Convolutional Neural Networks for Object Proposals and Detection[C]∥Applications of Computer Vision.IEEE,2017:924-933.
[2]HE K,ZHANG X,REN S,et al.Deep Residual Learning for Ima-
ge Recognition[C]∥IEEE Conference on Computer Vision and Pattern Recognition.IEEE Computer Society,2016:770-778.
[3]ROTH S,ANDRIYENKO A,SCHINDLER K.Discrete-conti-
nuous optimization for multi-target tracking[C]∥Computer Vision and Pattern Recognition.IEEE,2012:1926-1933.
[4]MILAN A,ROTH S,SCHINDLER K.Continuous optimization for multi-target tracking [J].IEEE transactions on pattern ana-lysis and machine intelligence,2014,36(1):58-72.
[5]ZHANG L,LI Y,NEVATIA R.Global data association for
multi-object tracking using network flows[C]∥IEEE Con-ference on Computer Vision and Pattern Recognition.IEEE,2008:1-8.
[6]SONG X,CUI J,ZHA H,et al.Vision-Based Multiple Interacting Targets Tracking via On-Line Supervised Learning[C]∥European Conference on Computer Vision.Springer-Verlag,2008:642-655.
[7]CHOI W,PANTOFARU C,SAVARESE S.A general framework for tracking multiple people from a moving camera[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(7):1577-1591.
[8]JU H Y,LEE C R,YANG M H,et al.Online Multi-object
Tracking via Structural Constraint Event Aggregation[C]∥Computer Vision and Pattern Recognition.IEEE,2016:1392-1400.
[9]CHOI W.Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor[C]∥IEEE International Conference on Computer Vision.IEEE,2015:3029-3037.
[10]BERCLAZ J,FLEURET F,TURETKEN E,et al.Multiple object tracking using k-shortest paths optimization[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(9):1806-1819.
[11]KALAL Z,MIKOLAJCZYK K,MATAS J.Forward-backward error:Automatic detection of tracking failures[C]∥2010 20th International Conference on Pattern recognition (ICPR).IEEE,2010:2756-2759.
[12]XIANG Y,ALAHI A,SAVARESE S.Learning to track:Online multi-object tracking by decision making[C]∥2015 IEEE International Conference on Computer Vision (ICCV).IEEE,2015:4705-4713.
[13]LI Y,HUANG C,NEVATIA R.Learning to associate:Hybridboosted multi-target tracker for crowded scene[C]∥IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009:2953-2960.
[14]HENRIQUES J F,RUI C,MARTINS P,et al.Exploiting the
Circulant Structure of Tracking-by-Detection with Kernels[C]∥European Conference on Computer Vision.Springer-Verlag,2012:702-715.
[15]HENRIQUES J F,RUI C,MARTINS P,et al.High-Speed
Tracking with Kernelized Correlation Filters[J].IEEE Transa-ctions on Pattern Analysis & Machine Intelligence,2015,37(3):583-596.
[16]BELLMAN R.A Markovian decision process[J].Journal of
Mathematics and Mechanics,1957,6(5):679-684.
[17]KALAL Z,MIKOLAJCZYK K,MATAS J.Tracking-learning-
detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(7):1409-1422.
[18]BERNARDIN K,STIEFELHAGEN R.Evaluating multiple object tracking performance:the CLEAR MOT metrics[OL].http://doi.org/10.1155/2008/246309.
[19]LI Y,HUANG C,NEVATIA R.Learning to associate:Hybridboosted multi-target tracker for crowded scene[C]∥IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009:2953-2960.
[20]GEIGER A,LENZ P,URTASUN R.Are we ready for autonomous driving? thekitti vision benchmark suite[C]∥IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2012:3354-3361.
[21]OEP A,MEHNER W,MATHIAS M,et al.Combined image-and world-space tracking in traffic scenes[C]∥IEEE International Conference on Robotics and Automation.2017:1988-1995.
[22]CHOI W.Near-online multi-target tracking with aggregated local flow descriptor[C]∥Proceedings of the IEEE International Conference on Computer Vision.IEEE,2015:3029-3037.
[23]ANDRIYENKO A,SCHINDLER K,ROTH S.Discrete-conti-
nuous optimization for multi-target tracking[C]∥2012 IEEE Conference onComputer Vision and Pattern Recognition.IEEE,2012:1926-1933.
[24]GEIGER A,LAUER M,WOJEK C,et al.3D Traffic Scene Understanding From Movable Platforms[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,36(5):1012-1025.
[25]ZHANG H,GEIGER A,URTASUN R.Understanding High-
Level Semantics by Modeling Traffic Patterns[C]∥IEEE International Conference on Computer Vision.IEEE,2013:3056-3063.
[26]XIANG Y,ALAHI A,SAVARESE S.Learning to track:Online multi-object tracking by decision making[C]∥Proceedings of the IEEE International Conference on Computer Vision.IEEE,2015:4705-4713.
[1] 沈祥培, 丁彦蕊.
多检测器融合的深度相关滤波视频多目标跟踪算法
Multi-detector Fusion-based Depth Correlation Filtering Video Multi-target Tracking Algorithm
计算机科学, 2022, 49(8): 184-190. https://doi.org/10.11896/jsjkx.210600004
[2] 文成宇, 房卫东, 陈伟.
多目标跟踪的对象初始化综述
Object Initialization in Multiple Object Tracking:A Review
计算机科学, 2022, 49(3): 152-162. https://doi.org/10.11896/jsjkx.210200048
[3] 刘彦, 秦品乐, 曾建朝.
基于YOLOv3与分层数据关联的多目标跟踪算法
Multi-object Tracking Algorithm Based on YOLOv3 and Hierarchical Data Association
计算机科学, 2021, 48(11A): 370-375. https://doi.org/10.11896/jsjkx.201000115
[4] 龚轩, 乐孜纯, 王慧, 武玉坤.
多目标跟踪中的数据关联技术综述
Survey of Data Association Technology in Multi-target Tracking
计算机科学, 2020, 47(10): 136-144. https://doi.org/10.11896/jsjkx.200200041
[5] 胡海根, 周莉莉, 周乾伟, 陈胜勇, 张俊康.
基于CNN的相衬显微图像序列的癌细胞多目标跟踪
Multi-target Tracking of Cancer Cells under Phase Contrast Microscopic Images Based
on Convolutional Neural Network
计算机科学, 2019, 46(5): 279-285. https://doi.org/10.11896/j.issn.1002-137X.2019.05.043
[6] 赵广辉, 卓松, 徐晓龙.
基于卡尔曼滤波的多目标跟踪方法
Multi-object Tracking Algorithm Based on Kalman Filter
计算机科学, 2018, 45(8): 253-257. https://doi.org/10.11896/j.issn.1002-137X.2018.08.045
[7] 袁大龙,纪庆革.
协同运动状态估计的多目标跟踪算法
Multiple Object Tracking Algorithm via Collaborative Motion Status Estimation
计算机科学, 2017, 44(Z11): 154-159. https://doi.org/10.11896/j.issn.1002-137X.2017.11A.032
[8] 陈金广,秦晓姗,马丽丽.
快速多目标跟踪GM-PHD滤波算法
Fast GM-PHD Filter for Multi-target Tracking
计算机科学, 2016, 43(3): 317-321. https://doi.org/10.11896/j.issn.1002-137X.2016.03.060
[9] 杨国亮,张进辉.
分层关联的多目标跟踪算法研究
Research on Multi-object Tracking Using Hierarchical Data Association
计算机科学, 2014, 41(9): 306-310. https://doi.org/10.11896/j.issn.1002-137X.2014.09.059
[10] 金鑫,梁雪春,袁晓龙.
复杂情况下的多目标跟踪统计技术
Multi-target Tracking Statistical Techniques in Complex Case
计算机科学, 2013, 40(6): 268-271.
[11] 黄镇谨,陆阳,杨娟,方欢.
Markov决策过程不确定策略特征模式
Property Patterns of Markov Decision Process Nondeterministic Choice Scheduler
计算机科学, 2013, 40(4): 263-266.
[12] 朱晓钢,杨兵,许华杰.
支持无线传感器网络多目标跟踪的聚类数据关联算法研究
Clustering Data Association Algorithm to Support Multi-target Tracl}ing in WSN
计算机科学, 2012, 39(Z6): 24-27.
[13] 周维,许海霞,郑金华.
基于RJMCMC的视觉多目标跟踪算法
Multi-object Visual Tracking Based on Reversible Jump Markov Chain Monte Carlo
计算机科学, 2012, 39(7): 270-275.
[14] 朱晓钢,杨兵,许华杰.
支持无线传感器网络多目标跟踪的最邻近数据关联算法研究
Nearest Neighbor Method Data Association Algorithm to Support Multi-target Tracking in WSN
计算机科学, 2011, 38(5): 67-70.
[15] 陈恒鑫 房斌 唐远炎 文静.
一种基于区域生长的多个形变目标跟踪方法

计算机科学, 2008, 35(4): 238-240.
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