基于注意力机制和孪生网络的跟踪算法研究

doi:10.3778/j.issn.1002-8331.2105-0140

摘要/Abstract

摘要： 提出融合卷积通道注意力机制、堆叠通道注意力机制和空间注意力机制的孪生网络跟踪器（ThrAtt-Siam）来提升跟踪性能。ThrAtt-Siam跟踪器以SiameseFC为基础，通过在低卷积层融合卷积通道注意力机制、两个特征图与两个卷积块，加强目标物体特征提取，提高跟踪器对背景特征抗干扰能力和辨别能力；在目标图像分支融合堆叠通道注意力机制与空间注意力机制，其中堆叠通道注意力机制可有效区分有用特征与无用特征，同时针对不同通道的有用特征进行提取，空间注意力机制可有效地补充目标物体特征在通道空间中的信息，能够更好地对目标进行定位。在OTB2015和VOT2017数据集上的实验结果表明，ThrAtt-Siam跟踪器对目标物体形变、低分辨率和遮挡问题都取得了较好的跟踪准确率和成功率。

关键词: 特征融合, 孪生网络, 注意力机制, 目标跟踪

Abstract: This paper proposes a twin network tracker which combines convolutional channel attention mechanism, stacked channel attention mechanism and spatial attention mechanism to improve tracking performance（ThrAtt-Siam）. ThrAtt-Siam tracker is based on SiameseFC. By fusing convolution channel attention mechanism, two feature maps and two convolution blocks in the low convolution layer, the target feature extraction is enhanced, and the anti-interference ability and discrimination ability of the tracker against background features are improved. The stackable channel attention mechanism and spatial attention mechanism are fused in the target image branch. The stackable channel attention mechanism can effectively distinguish useful features from useless features. At the same time, the useful features of different channels are extracted. The spatial attention mechanism can effectively supplement the information of the target object features in the channel space, and can better locate the target. The experimental results on OTB2015 and VOT2017 data sets show that ThrAtt-Siam tracker achieves good tracking accuracy and success rate for target deformation, low resolution and occlusion problems.

Key words: feature fusion, twin network, attention mechanism, target tracking

王玲, 周磊, 王鹏, 白燕娥. 基于注意力机制和孪生网络的跟踪算法研究[J]. 计算机工程与应用, 2022, 58(23): 161-168.

WANG Ling, ZHOU Lei, WANG Peng, BAI Yan’e. Research on Tracking Algorithm Based on Attention Mechanism and Siamese Network[J]. Computer Engineering and Applications, 2022, 58(23): 161-168.

参考文献

[1] YUN S，CHOI J，YOO Y，et al.Action-decision networks for visual tracking with deep reinforcement learning[C]//IEEE Conference on Computer Vision and Pattern Recog-nition，2017：1349-1358.
[2] GUO Q，WEI F，ZHOU C，et al.Learning dynamic siamese network for visual object tracking[C]//2017 IEEE International Conference on Computer Vision（ICCV），2017：1781-1789.
[3] HE A，CHONG L，TIAN X，et al.A twofold siamese network for real-time object tracking[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition，2018：4834-4843.
[4] BHAT G，DANELLJAN M，GOOL L V，et al.Learning discriminative model prediction for tracking[C]//2019 IEEE/CVF International Conference on Computer Vision，2019：6181-6190.
[5] ZHENG L，TANG M，CHEN Y，et al.Siamese deformable cross-correlation network for real-time visual tracking[J].Neurocomputing，2020，401：36-47.
[6] BERTINETTO L，VALMADRE J.Fully-convolutional siamese networks for object tracking[C]//European Conference on Computer Vision.Cham：Springer，2016：850-865.
[7] 李睿，连继荣.改进的Siamese自适应网络和多特征融合跟踪算法[J].计算机科学与探索，2022，16（11）：2587-2595.
LI R，LIAN J R.Improved Siamese adaptive network and multi-feature fusion tracking algorithm[J].Journal of Frontiers of Computer Science and Technology，2022，16（11）：2587-2595.
[8] 王希鹏，李永，李智，等.引入短时记忆的Siamese网络目标跟踪算法[J].计算机工程与应用，2022，58（3）：235-241.
WANG X P，LI Y，LI Z，et al.Siamese network tracking with short-term memory[J].Computer Engineering and Applications，2022，58（3）：235-241.
[9] WANG Q，TENG Z，XING J，et al.Learning attentions：residual attentional siamese network for high performance online visual tracking[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition，2018：4854-4863.
[10] 张汉，张德祥，陈鹏，等.并行注意力机制在图像语义分割中的应用[J].计算机工程与应用，2022，58（9）：151-160.
ZHANG H，ZHANG D X，CHEN P，et al.Application of parallel attention mechanism in image semantic segmentation[J].Computer Engineering and Applications，2022，58（9）：151-160.
[11] YANG K，HE Z，ZHOU Z，et al.SiamAtt：siamese attention network for visual tracking[J].Knowledge-Based Systems，2020，203：106079.
[12] WANG Q，WU B，ZHU P，et al.ECA-Net：efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition（CVPR），2020：11531-11539.
[13] 刘满，胡磊，宁纪锋，等.基于原型注意力的多域网络目标跟踪方法[J].计算机工程与应用，2022，58（20）：206-211.
LIU M，HU L，NING J F.et al.Multi-domain network target tracking method based on prototype attention[J].Computer Engineering and Applications，2022，58（20）：206-211.
[14] TANG C，QIN P，ZHANG J.Robust template adjustment siamese network for object visual tracking[J].Sensors，2021，21（4）：1466.
[15] FIAZ M，MAHMOOD A，JUNG S K.Learning soft mask based feature fusion with channel and spatial attention for robust visual object tracking[J].Sensors，2020，20（14）：4021.
[16] RAHMAN M M，AHED M R，LAISHRAM L，et al.Siamese high-level feature refine network for visual object tracking[J].Electronics，2020，9（11）：1918.
[17] ZHU Z，HUANG G，ZOU W，et al.UCT：learning unified convolutional networks for real-time visual tracking[C]//2017 IEEE International Conference on Computer Vision Workshops（ICCVW），2017：1973-1982.
[18] BERTINETTO L，VALMADRE J，GOLODETZ S，et al.Staple：complementary learners for real-time tracking[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition，2016：1401-1409.
[19] 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.
[20] DANELLJAN M，HAGER G，KHAN F S，et al.Discrimi-native scale space tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2017，39（8）：1561-1575.
[21] GALOOGAHI H K，FAGG A，LUCEY S.Learning background-aware correlation filters for visual tracking[C]//2017 IEEE International Conference on Computer Vision，2017：1144-1152.
[22] WANG M，LIU Y，HUANG Z.Large margin object tracking with circulant feature maps[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition，2017：4800-4808.