Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2021, Vol. 55 Issue (10): 1834-1846    DOI: 10.3785/j.issn.1008-973X.2021.10.005
计算机技术     
自适应上下文感知的目标追踪方法
柏昀旭1(),陆新江1,*(),骆锐2
1. 中南大学 机电工程学院，湖南 长沙 410083
2. 湖南工业职业技术学院，湖南 长沙 410083
Adaptive context-aware target tracking method
Yun-xu BAI1(),Xin-jiang LU1,*(),Rui LUO2
1. College of Mechanical and Electrical Engineering, Central South University, Changsha 410083, China
2. Hunan Industry Polytechnic, Changsha 410083, China
 全文: PDF(3165 KB)   HTML
摘要：

针对相关滤波方法容易受到背景干扰导致跟踪漂移的问题，提出自适应上下文感知图像跟踪方法. 为了减少背景干扰，选取离目标位置远的高响应区域为自适应上下文区域，赋予该区域自适应的低响应值. 根据上下文区域与目标区域响应的相对差值，给上下文区域自适应的惩罚因子，使得该算法具有更好的鲁棒性. 该算法在OTB2013、OTB2015及Temple-Color128标准数据集上都展现了优秀的跟踪性能，OTB2015的重叠率精度达到61.53%，超过大部分已有的优秀算法，特别是在背景混叠及部分遮挡的情况下有着更卓越的表现. 该算法的平均跟踪速度为24.5帧/s，实时性较好.
关键词： 相关滤波;  目标追踪;  自适应上下文感知;  背景干扰;  跟踪漂移    
Abstract:

An adaptive context-aware target tracking method was proposed aiming at the problem that the correlation filtering methods were easily interfered by background, which led to tracking drift. The high response area far from the target position was selected as the adaptive context area in order to reduce the background interference. Then the adaptive low response value was assigned to the area. The penalty factor was adaptively given to the context area according to the relative difference of response value between the context area and the target area, which made the algorithm more robust. The algorithm showed excellent tracking performance on OTB2013, OTB2015 and Temple-Color128 benchmark. The overlapping rate accuracy of OTB2015 was 61.53%, which was superior to most existing excellent algorithms. The algorithm performed better especially in the case of background clutter and partial occlusion. The average tracking speed of the algorithm was 24.5 frames per second, and the algorithm had a good real-time effect.
Key words: correlation filtering    target tracking    adaptive context-aware    background interference    tracking drift
收稿日期: 2020-11-03 出版日期: 2021-10-27
CLC:  TP 391  
基金资助: 国家重点研发计划资助项目（2018YFB1308202）；国家自然科学基金资助项目（52075556）；湖南省杰青资助项目（2019JJ20030）；湖南省高新技术产业科技创新引领计划资助项目（2020GK4097）；湖南省教育厅科学研究资助项目（12B035）
通讯作者: 陆新江     E-mail: yxbai2017@csu.edu.cn;xjlu@csu.edu.cn
作者简介: 柏昀旭（1995—），男，博士生，从事机器学习与目标跟踪的研究. orcid.org/0000-0003-1601-7184. E-mail： yxbai2017@csu.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
柏昀旭
陆新江
骆锐

引用本文:

柏昀旭,陆新江,骆锐. 自适应上下文感知的目标追踪方法[J]. 浙江大学学报(工学版), 2021, 55(10): 1834-1846.

Yun-xu BAI,Xin-jiang LU,Rui LUO. Adaptive context-aware target tracking method. Journal of ZheJiang University (Engineering Science), 2021, 55(10): 1834-1846.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2021.10.005    或    https://www.zjujournals.com/eng/CN/Y2021/V55/I10/1834

图 1  传统上下文感知的示意图
图 2  上下文区域的自适应选取思路
图 3  上下文区域响应值
图 4  自适应惩罚因子
图 5  自适应上下文感知算法的流程图
图 6  不同K下提出算法与传统CACF方法的DP精度和AUC成功率对比
算法 DP /% OP /% AUC /% v /（帧·s?1）
ACACF（K=1） 85.72 72.63 60.39 25.8
ACACF（K=2） 86.87 74.74 61.53 24.5
ACACF（K=3） 85.43 72.21 60.10 23.8
ACACF（K=4） 79.73 67.79 56.81 21.8
CACF 80.11 70.24 58.56 25.7
表 1  不同K下的各评价指标得分
算法 DP /% OP /% AUC /% v /（帧·s?1）
ADA_CA（K=2） 88.16 81.12 64.29 27.2
BACF 83.87 81.41 63.78 27.1
ARCF 82.84 79.18 62.62 4.5
Staple 79.26 75.39 59.95 44.9
SRDCF 83.79 78.13 62.62 4.5
CSRDCF 80.28 73.78 59.33 22.8
LMCF 84.20 80.03 62.76 77.6
LCT 84.12 80.70 62.37 21.6
SAMF 78.50 73.19 57.93 18.6
fDSST 73.97 67.03 55.42 19.4
CFNet 78.49 75.19 58.89 NaN
SiamFC 80.93 77.86 60.73 NaN
DCFNet 79.45 77.86 62.24 21.2
ACFN 85.96 75.03 60.71 7.0
HCF 87.86 73.22 59.75 8.0
HDT 87.74 72.96 59.64 4.3
表 2  优秀跟踪算法在OTB2013上的各评价指标得分
图 7  提出算法与优秀跟踪算法在OTB2013数据集上的DP精度和AUC成功率对比
算法 DP /% OP /% AUC /% v /（帧·s?1）
ADA_CA（K=2） 86.87 76.34 61.53 24.5
BACF 81.06 76.74 61.09 27.5
ARCF 80.65 74.65 60.66 24.4
Staple 78.40 70.92 58.13 42.9
SRDCF 78.95 72.84 59.80 4.3
CSRDCF 80.24 70.19 58.69 22.6
LMCF 78.85 71.88 58.01 42.9
LCT 75.84 69.77 55.95 20.7
SAMF 75.13 67.37 55.32 17.0
fDSST 68.67 60.53 51.73 17.7
CFNet 77.71 73.67 58.62 NaN
SiamFC 77.05 73.05 58.21 NaN
DCFNet 75.07 41.20 57.99 41.2
ACFN 79.90 69.25 57.31 10.0
HCF 83.10 65.14 55.81 10.4
HDT 84.20 65.35 56.11 5.5
表 3  优秀跟踪算法在OTB2015数据集上的各评价指标得分
图 8  提出算法与优秀跟踪算法在OTB2015数据集上的DP精度和AUC成功率对比
算法 DP /% OP /% AUC /% v /（帧·s?1）
ADA_CA（K=2） 72.81 66.01 53.41 21.1
BACF 64.58 61.30 48.64 37.6
ARCF 70.27 64.59 51.94 22.2
Staple 66.66 62.01 49.71 83.2
SRDCF 68.91 61.30 50.53 2.8
CSRDCF 67.74 59.19 50.04 16.6
MEEM 70.87 61.16 49.77 20.5
LCT 60.59 52.56 43.17 39.3
SAMF 63.07 57.88 46.56 27.1
fDSST 53.34 46.95 40.50 16.3
RCT 72.60 65.60 52.95 34.3
WSCF 69.98 63.90 51.08 24.5
TRACF 71.33 64.81 52.21 23.7
TSC 72.09 66.50 53.34 2.2
DAMA 67.69 61.01 49.72 30.4
PSCA 71.03 63.23 51.33 1.8
表 4  优秀跟踪算法在TC128数据集上的各评价指标得分
图 9  提出算法与优秀跟踪算法在TC128数据集上的DP精度和AUC成功率对比
图 10  在复杂的视频序列上可视化的跟踪结果（Human3，Girl，Soccer，DragonBaby，iroman）
图 11  优秀跟踪算法在OTB-2015 11个属性上的成功率曲线图
图 12  跟踪失败案例（carScale，freeman4，matrix，motorRolling）
1 MOROZOV A Development of a method for intelligent video monitoring of abnormal behavior of people based on parallel object-oriented logic programming[J]. Pattern Recognition and Image Analysis, 2015, 25 (3): 481- 492
doi: 10.1134/S1054661815030153
2 CHEN D, BHARUCHA A J, WACTLAR H D. Intelligent video monitoring to improve safety of older persons [C]// Proceedings of the 29th Annual International Conference of the IEEE-Engineering-in-Medicine-and-Biology-Society. Lyon: IEEE, 2007: 3814-3817.
3 CHANG C-K, SIAGIAN C, ITTI L. Mobile robot vision navigation and localization using gist and saliency [C]// Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems. Taipei: IEEE, 2010: 4147-4154.
4 王金川, 黄雪梅, 郝园 一种改进的医学图像目标轮廓跟踪算法[J]. 现代制造工程, 2012, 10: 33- 36
WANG Jin-chuan, HUANG Xue-mei, HAO Yuan An improved contour tracking algorithm for medical images[J]. Modern Manufacturing Engineering, 2012, 10: 33- 36
doi: 10.3969/j.issn.1671-3133.2012.03.010
5 赵怀勋, 姜莹礁 基于颜色和方向梯度特征融合的粒子滤波跟踪算法[J]. 计算机应用与软件, 2014, 31 (6): 260- 263
ZHAO Huai-xun, JIANG Ying-jiao Particle filter tracking algorithm based on color and direction gradient feature fusion[J]. Computer Applications and Software, 2014, 31 (6): 260- 263
doi: 10.3969/j.issn.1000-386x.2014.06.070
6 卫保国, 赵思同, 文绪亮 基于局部直方图的多区域目标跟踪算法[J]. 计算机工程与应用, 2018, 54 (14): 26- 33
WEI Bao-guo, ZHAO Si-tong, WEN Xu-liang Multi-region target tracking algorithm based on local histogram[J]. Computer Engineering and Application, 2018, 54 (14): 26- 33
doi: 10.3778/j.issn.1002-8331.1805-0083
7 HENRIQUES J F, CASEIRO R, MARTINS P, et al High-speed tracking with kernelized correlation filters[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37 (3): 583- 596
doi: 10.1109/TPAMI.2014.2345390
8 YANG L, ZHU J. A scale adaptive kernel correlation filter tracker with feature integration [C]// Proceedings of the 13th European Conference on Computer Vision. Zurich: Springer, 2014: 254-265.
9 BERTINETTO L, VALMADRE J, GOLODETZ S, et al. Staple: complementary learners for real-time tracking [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2016: 1401-1409.
10 DANELLJAN M, HAGER G, SHAHBAZ KHAN F, et al. Convolutional features for correlation filter based visual tracking [C]// Proceedings of the IEEE International Conference on Computer Vision. Santiago: IEEE, 2015: 58-66.
11 MA C, HUANG J B, YANG X, et al. Hierarchical convolutional features for visual tracking[C]// Proceedings of the IEEE International Conference on Computer Vision. Santiago: IEEE, 2016.
12 CHOI J, KWON J, LEE K M. Deep meta learning for real-time target-aware visual tracking [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 911-920.
13 XU T, FENG Z, WU X, et al. Joint group feature selection and discriminative filter learning for robust visual object tracking [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 7949-7959.
14 TANG M, FENG J. Multi-kernel correlation filter for visual tracking [C]// Proceedings of the IEEE International Conference on Computer Vision. Santiago: IEEE, 2016: 3038-3046.
15 BHAT G, DANELLJAN M, VAN GOOL L, et al. Learning discriminative model prediction for tracking [C]// proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 6181-6190.
16 LI P, CHEN B, OUYANG W, et al. GradNet: gradient-guided network for visual object tracking [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 6161-6170.
17 GALOOGAHI H K, FAGG A, LUCEY S. Learning background-aware correlation filters for visual tracking [C]// Proceedings of the 16th IEEE International Conference on Computer Vision. Venice: IEEE, 2017: 1144-1152.
18 GALOOGAHI H K, SIM T, LUCEY S. Correlation filters with limited boundaries [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston: IEEE, 2015: 4630-4638.
19 DANELLJAN M, HÄGER G, KHAN F S, et al. Learning spatially regularized correlation filters for visual tracking [C]// Proceedings of the IEEE International Conference on Computer Vision. Santiago: IEEE, 2015: 4310-4318.
20 MUELLER M, SMITH N, GHANEM B. Context-aware correlation filter tracking [C]// Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 1387-1395.
21 LUKEI A, VOJÍ T, EHOVINZAJC L, et al. Discriminative correlation filter with channel and spatial reliability [C]// Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2018: 4847-4856.
22 HUANG Z, FU C, LI Y, et al. Learning aberrance repressed correlation filters for real-time UAV tracking [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 2891-2900.
23 LI Y, ZHU J. A scale adaptive kernel correlation filter tracker with feature integration [C]// Proceedings of the 13th European Conference on Computer Vision. Zurich: Springer, 2014: 254-265.
24 DANELLJAN M, HÄGER G, KHAN F S, et al. Accurate scale estimation for robust visual tracking [C]// Proceedings of the British Machine Vision Conference. Nottingham: [s.n.], 2014: 1-5.
25 DANELLJAN M, HAGER G, KHAN F S, et al Discriminative scale space tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39 (8): 1561- 1575
doi: 10.1109/TPAMI.2016.2609928
26 WANG X, HOU Z, YU W, et al Robust occlusion-aware part-based visual tracking with object scale adaptation[J]. Pattern Recognition, 2018, 81: 456- 470
doi: 10.1016/j.patcog.2018.04.011
27 GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation [C]// Proceedings of the 27th IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 2014: 580-587.
28 QI Y, ZHANG S, QIN L, et al. Hedged deep tracking [C]// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2016: 4303-4311.
29 VALMADRE J, BERTINETTO L, HENRIQUES J, et al. End-to-end representation learning for correlation filter based tracking [C]// Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 5000-5008.
30 BERTINETTO L, VALMADRE J, HENRIQUES J F, et al. Fully-convolutional siamese networks for object tracking [C]// Proceedings of the 14th European Conference on Computer Vision. Amsterdam: Springer, 2016: 850-865.
31 DANELLJAN M, ROBINSON A, KHAN F S, et al. Beyond correlation filters: learning continuous convolution operators for visual tracking [C]// Proceedings of the 14th European Conference on Computer Vision. Amsterdam: Springer, 2016: 472-488.
32 HUANG L, ZHAO X, HUANG K, et al. Bridging the gap between detection and tracking: a unified approach [C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul: IEEE, 2019: 3998-4008.
33 WU Y, LIM J, YANG M-H, et al. Online object tracking: a benchmark [C]// Proceedings of the 26th IEEE Conference on Computer Vision and Pattern Recognition. Portland: IEEE, 2013: 2411-2418.
34 WU Y, LIM J, YANG M-H Object tracking benchmark[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37 (9): 1834- 1848
doi: 10.1109/TPAMI.2014.2388226
35 LIANG P, BLASCH E, LING H Encoding color information for visual tracking: algorithms and benchmark[J]. IEEE Transactions on Image Processing, 2015, 24 (12): 5630- 5644
doi: 10.1109/TIP.2015.2482905
36 WANG M, LIU Y, HUANG Z, et al. Large margin object tracking with circulant feature maps [C]// Proceedings of the 30th IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 4800-4808.
37 MA C, YANG X, ZHANG C, et al. Long-term correlation tracking [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston: IEEE, 2015: 5388-5396.
38 ZHANG J, MA S, SCLAROFF S. MEEM: robust tracking via multiple experts using entropy minimization [C]// Proceedings of the 13th European Conference on Computer Vision. Zurich: Springer, 2014: 188-203.
39 WANG Q, GAO J, XING J, et al. DCFNet: discriminant correlation filters network for visual tracking [EB/OL]. [2017-04-13]. http://arxiv.org/abs/1704.04057.
40 CHOI J, CHANG H J, YUN S, et al. Attentional correlation filter network for adaptive visual tracking [C]// Proceedings of the 30th IEEE/CVF Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 4828-4837.
41 LI D, PORIKLI F, WEN G, et al When correlation filters meet siamese networks for real-time complementary tracking[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30 (2): 509- 519
doi: 10.1109/TCSVT.2019.2892759
42 HAN R, FENG W, WANG S Fast learning of spatially regularized and content aware correlation filter for visual tracking[J]. IEEE Transactions on Image Processing, 2020, 29: 7128- 7140
doi: 10.1109/TIP.2020.2998978
43 ZHOU Y, LI J, DU B, et al A target response adaptive correlation filter tracker with spatial attention[J]. Multimedia Tools And Applications, 2020, 79 (29/30): 20521- 20543
44 WANG H, WANG Z, FANG B, et al Temporal–spatial consistency of self-adaptive target response for long-term correlation filter tracking[J]. Signal, Image and Video Processing, 2020, 14 (4): 639- 644
doi: 10.1007/s11760-019-01594-2
45 CHEN G, PAN G, ZHOU Y, et al Correlation filter tracking via distractor-aware learning and multi-anchor detection[J]. IEEE Transactions on Circuits And Systems for Video Technology, 2020, 30 (12): 4810- 4822
doi: 10.1109/TCSVT.2019.2961999
[1] 熊昌镇,卢颜,闫佳庆. 基于各向异性高斯分布的视觉跟踪算法[J]. 浙江大学学报(工学版), 2020, 54(2): 301-310.
[2] 熊昌镇,王润玲,邹建成. 基于多高斯相关滤波的实时跟踪算法[J]. 浙江大学学报(工学版), 2019, 53(8): 1488-1495.