基于注意力学习的正则化相关滤波跟踪算法

doi:10.3969/j.issn.0372-2112.2020.09.014

电子学报 ›› 2020, Vol. 48 ›› Issue (9): 1762-1768.DOI: 10.3969/j.issn.0372-2112.2020.09.014

基于注意力学习的正则化相关滤波跟踪算法

仇祝令¹, 查宇飞^1,2, 吴敏³, 王青³

1. 中国人民解放军空军工程大学航空工程学院, 陕西西安 710038;
2. 西北工业大学计算机学院, 陕西西安 710072;
3. 95894 部队, 北京 102211

收稿日期:2019-12-26 修回日期:2020-01-26 出版日期:2020-09-25
- 通讯作者:
- 查宇飞
- 作者简介:
- 仇祝令 男,1995年出生,山东人.空军工程大学在读硕士研究生,主要研究方向:目标跟踪、机器学习以及计算机视觉.
- 基金资助:
- 国家自然科学基金 (No.61472442，No.61773397，No.61701524）; 航空科学基金 (No.2020-HT-XG）; 中央高校基本科研业务费专项资金资助 (No.3102019ZY1003，No.3102019ZY1004）

Learning Attentional Regularized Correlation Filter for Visual Tracking

QIU Zhu-ling¹, ZHA Yu-fei^1,2, WU Min³, WANG Qing³

1. Aeronautics Engineering College, Air Force Engineering University, Xi'an, Shaanxi 710038, China;
2. School of Computer Science, Northwestern Polytechnical University, Xi'an, Shaanxi 710072, China;
3. Unit 95972 of the PLA, Beijing 102211, China

Received:2019-12-26 Revised:2020-01-26 Online:2020-09-25 Published:2020-09-25

摘要/Abstract

摘要： 边界效应是制约相关滤波跟踪性能的一个重要因素.目前大多数方法只是简单地采用先验知识，如逆高斯分布，预设掩模等，或者分割前景目标作为正则化项，进行约束求解，并没有考虑目标的空时域特性.针对这一问题，本文提出一种基于注意力学习的正则化相关滤波跟踪算法.该方法考虑目标在空间中的分布特性，利用注意力机制学习目标的特定空间权重，适应目标在空域中的变化；同时利用目标在时域中的连续性，通过对注意力权重矩阵的约束来间接调整滤波器；最后通过交替方向乘子（ADMM）算法迭代优化模型.我们在标准的数据库上进行大量实验，结果表明本文算法能实时跟踪目标，并且在精确度和成功率上都有了一定的提升.

关键词: 单目标, 视觉跟踪, 机器学习, 正则化, 相关滤波, 注意力学习

Abstract: Boundary effect is an important factor which restricts the performance of correlation filter.At present,most methods simply use the prior knowledge,such as inverse Gaussian distribution,preset masks,etc.,or segment the foreground target to constrain solving as the regularization term,which do not consider characteristics of the target in the spatial and temporal domain.To address this problem,this paper proposes a learning attention regularized correlation filter for visual tracking.The method uses the attention mechanism to learn the specific spatial weight of the target,which can adapt to the variations of target in the spatial domain by considering the spatial distribution characteristics of the target.At the same time,this paper uses the continuity of the target in the temporal domain.The filter is indirectly adjusted by constraining the attention weight matrix.Finally,the alternating direction method of multipliers (ADMM) is employed to iteratively optimize the model.We conduct extensive experiments on the proposed method in the standard tracking database.The results show that the proposed algorithm can track the target in real time,and has a certain improvement in precision and success rate.

Key words: single target, visual tracking, machine learning, regularization, correlation filter, attention mechanism

中图分类号:

TN919.82

仇祝令, 查宇飞, 吴敏, 王青. 基于注意力学习的正则化相关滤波跟踪算法[J]. 电子学报, 2020, 48(9): 1762-1768.

QIU Zhu-ling, ZHA Yu-fei, WU Min, WANG Qing. Learning Attentional Regularized Correlation Filter for Visual Tracking[J]. Acta Electronica Sinica, 2020, 48(9): 1762-1768.

参考文献

[1] 孙航,李晶,杜博,等.基于多阶段学习的相关滤波目标跟踪[J].电子学报,2017,45(10):2337-2342. Sun Hang,Li Jing,Du Bo,et al.Correlation filter target tracking based on multi-stage learning[J].Acta Electronica Sinica,2017,45(10):2337-2342.(in Chinese)
[2] 范舜奕,管桦,侯志强,等.基于多表观特征子模型更新的鲁棒视觉跟踪[J].电子学报,2018,46(02):440-446. Fan Shun-yi,Guan Hua,Hou Zhi-qiang,et al.Robust visual tracking based on multi-epitopic characterization sub-model updating[J].Acta Electronica Sinica,2018,46(02):440-446.(in Chinese)
[3] 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,2014,37(3):583-596.
[4] Wu Y,Lim J,Yang M H.Object tracking benchmark[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9):1834-1848.
[5] Kristan M,Leonardis A,Matas J,et al.The visual object tracking vot2017 challenge results[A].Proceedings of the International Conference on Computer Vision[C].Venice:IEEE,2017.1949-1972.
[6] Danelljan M,Hager G,Khan F S,et al.Learning spatially regularized correlation filters for visual tracking[A].Proceedings of the International Conference on Computer Vision[C].Santiago:IEEE,2015.4310-4318.
[7] Kiani Galoogahi H,Fagg A,Lucey S.Learning background-aware correlation filters for visual tracking[A].Proceedings of the International Conference on Computer Vision[C].Venice:IEEE,2017.1135-1143.
[8] Lukezic A,Vojir T,?ehovin Zajc L,et al.Discriminative correlation filter with channel and spatial reliability[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Honolulu,HI:IEEE,2017.6309-6318.
[9] Ghadimi E,Teixeira A,Shames I,et al.Optimal parameter selection for the alternating direction method of multipliers (ADMM):quadratic problems[J].IEEE Transactions on Automatic Control,2014,60(3):644-658.
[10] Li F,Tian C,Zuo W,et al.Learning spatial-temporal regularized correlation filters for visual tracking[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Las Vegas:IEEE,2018.4904-4913.
[11] Pu S,Song Y,Ma C,et al.Deep attentive tracking via reciprocative learning[A].Advances in Neural Information Processing Systems[C].Montréal:Curran Associates Inc,2018.1931-1941.
[12] Choi J,Jin Chang H,Yun S,et al.Attentional correlation filter network for adaptive visual tracking[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Honolulu,HI:IEEE,2017.4807-4816.
[13] Bertinetto L,Valmadre J, Henriques J F,et al.Fully-convolutional siamese networks for object tracking[A].Proceedings of the European Conference on Computer Vision[C].Amsterdam:Springer,Cham,2016.850-865.
[14] Danelljan M,Häger G,Khan F S,et al.Discriminative scale space tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2016,39(8):1561-1575.
[15] Danelljan M,Bhat G,Shahbaz Khan F,et al.Eco:Efficient convolution operators for tracking[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].Honolulu,HI:IEEE,2017.6638-6646.

基于注意力学习的正则化相关滤波跟踪算法

Learning Attentional Regularized Correlation Filter for Visual Tracking

PDF

可视化

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

[1]	李雪莹, 王田路, 梁鹏, 王翀. 基于系统模型的用户评论中非功能需求的自动分类[J]. 电子学报, 2022, 50(9): 2079-2089.
[2]	韩亮, 蔡文涛, 蒲秀娟, 罗统军, 黄谦. 使用信噪比正则化LightGBM的腹部源胎儿心电信号提取方法[J]. 电子学报, 2022, 50(8): 1793-1800.
[3]	江兵兵, 何文达, 吴兴宇, 项俊浩, 洪立斌, 盛伟国. 基于自适应图学习的半监督特征选择[J]. 电子学报, 2022, 50(7): 1643-1652.
[4]	黄璐, 蔚保国, 李宏生, 李隽, 贾浩男, 程建强, 李雅宁. GNSS拒止环境下的伪卫星指纹定位方法[J]. 电子学报, 2022, 50(4): 811-822.
[5]	崔亚奇, 何友, 唐田田, 熊伟. 一种深度学习航迹关联方法[J]. 电子学报, 2022, 50(3): 759-763.
[6]	杨晓城, 杨真乙, 阎敬业, 武林, 蒋明峰. 基于拉普拉斯算子的综合孔径辐射计成像算法[J]. 电子学报, 2022, 50(2): 339-345.
[7]	蒋华伟, 郭陶, 杨震. 车辆路径问题研究进展[J]. 电子学报, 2022, 50(2): 480-492.
[8]	郑锐, 汪秋云, 林卓庞, 靖蓉琦, 姜政伟, 傅建明, 汪姝玮. 一种基于威胁情报层次特征集成的挖矿恶意软件检测方法[J]. 电子学报, 2022, 50(11): 2707-2715.
[9]	刘浩阳, 林耀进, 刘景华, 吴镒潾, 毛煜, 李绍滋. 由粗到细的分层特征选择[J]. 电子学报, 2022, 50(11): 2778-2789.
[10]	张梦, 郑建宏, 刘香燕, 何云. 基于集成学习的全双工中继系统安全中继选择方案研究[J]. 电子学报, 2021, 49(9): 1852-1856.
[11]	郑海颖, 王峰, 姜维, 王志强, 姚西文. 神经网络训练策略对高分辨率遥感图像场景分类性能影响的评估[J]. 电子学报, 2021, 49(8): 1599-1614.
[12]	周东明, 张灿龙, 李志欣, 王智文. 基于多层级视觉融合的图像描述模型[J]. 电子学报, 2021, 49(7): 1286-1290.
[13]	侯秀聪, 赖晓平, 曹九稳. 正则化超限学习机的最大分划广义交替方向乘子法[J]. 电子学报, 2021, 49(4): 625-630.
[14]	陈丹, 姚伯羽. 运动模型引导的自适应核相关目标跟踪方法[J]. 电子学报, 2021, 49(3): 550-558.
[15]	陈长青, 郭春, 崔允贺, 申国伟, 蒋朝惠. 基于API短序列的勒索软件早期检测方法[J]. 电子学报, 2021, 49(3): 586-595.