Abstract：Aiming at the problem that the tracking target is blocked or there are sensitive interferences around the target, which leads to the classification error of foreground background and the prediction error of boundary box, an anchor frame-free target tracking method based on cross-channel attention is proposed. Firstly, cross-channel attention is used to enhance the last three layers of output of feature extraction, and the correlation of all channel features is integrated by using the similarity of target in template features and search features, so as to selectively enhance the channel of target features. After that, feature fusion is carried out by weighted summation, and shallow feature fusion and deep feature fusion are used to improve classification accuracy and location accuracy. Finally, location attention is used to enhance the features of the classification feature map again to improve the accuracy of target location. Experimental results show that the proposed algorithm achieves 85.5% precision rate and 64.1% success rate on OTB100 dataset, and 70.5% precision rate and 56.0% success rate on UAV20L dataset.
Gao J, Zhang T, Xu C. Graph convolutional tracking ［C］//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2019: 4649-4659.
Kong D M, Yang D, Wang S T. Research on interactive Multi-model set adaptive collaborative filtering target tracking Algorithm ［J］. Acta Metrologica Sinica, 2021, 42(5): 638-644.
［3］
Possegger H, Mauthner T, Bischof H. In defense of color-based model-free tracking ［C］//IEEE conference on computer vision and pattern recognition. 2015: 2113-2120.
［13］
He K, Zhang X, Ren S, et al. Deep residual learning for image recognition ［C］//IEEE conference on computer vision and pattern recognition. 2016: 770-778.
Chen Z W, Zhang Z X, Song J, et al. Tracking Algorithm for Siamese Network Based on Target-Aware Feature Selection ［J］. Acta Optica Sinica, 40(9): 110-126.
［19］
Zhao H, Shi J, Qi X, et al. Pyramid scene parsing network ［C］// IEEE conference on computer vision and pattern recognition. 2017: 2881-2890.
［4］
Valmadre J, Bertinetto L, Henriques J, et al. End-to-end representation learning for correlation filter based tracking ［C］//IEEE conference on computer vision and pattern recognition. 2017: 2805-2813.
［6］
Zhang L, Varadarajan J, Nagaratnam Suganthan P, et al. Robust visual tracking using oblique random forests ［C］//IEEE conference on computer vision and pattern recognition. 2017: 5589-5598.
［9］
Ren S, He K, Girshick R, et al. Faster r-cnn: Towards real-time object detection with region proposal networks ［J］. Advances in neural information processing systems, 2015, 28: 1137-1149.
［10］
Li B, Yan J, Wu W, et al. High performance visual tracking with siamese region proposal network ［C］// IEEE conference on computer vision and pattern recognition. 2018: 8971-8980.
［11］
Li B, Wu W, Wang Q, et al. Siamrpn++: Evolution of siamese visual tracking with very deep networks ［C］//IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019: 4282-4291.
［14］
Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks ［J］. Advances in neural information processing systems, 2012, 25: 89-90.
［18］
Zhang Z, Lan C, Zeng W, et al. Relation-aware global attention for person re-identification ［C］// IEEE/CVF conference on computer vision and pattern recognition. 2020: 3186-3195.
［21］
Huang L, Zhao X, Huang K. Got-10k: A large high-diversity benchmark for generic object tracking in the wild ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 43(5): 1562-1577.
［22］
Muller M, Bibi A, Giancola S, et al. Trackingnet: A large-scale dataset and benchmark for object tracking in the wild ［C］//European conference on computer vision. 2018: 300-317.
Zhang L G, Yang M, Zhou S E, et al. Multi-scale Correlation Filtering Tracking Based on Adaptive Feature Fusion ［J］. Acta Metrologica Sinica, 2022, 43(10): 1271-1278.
［2］
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.
［7］
Danelljan M, Bhat G, Khan F S, et al. Accurate tracking by overlap maximization ［C］//IEEE/CVF conference on computer vision and pattern recognition. 2019: 4660-4669.
［8］
Bertinetto L, Valmadre J, Henriques J F, et al. Fully-convolutional siamese networks for object tracking ［C］//European conference on computer vision. Springer, Cham, 2016: 850-865.
［12］
Zhang Z, Peng H. Deeper and wider siamese networks for real-time visual tracking ［C］//IEEE/CVF conference on computer vision and pattern recognition. 2019: 4591-4600.
Chen W D, Chen L, Deng Z W, et al. Object tracking algorithm based on information fusion and redetection of hybrid correlation filtering ［J］. Acta Metrologica Sinica, 2019, 40(6): 1006-1012.
Fu J, Liu J, Tian H, et al. Dual attention network for scene segmentation ［C］// IEEE/CVF conference on computer vision and pattern recognition. 2019: 3146-3154.
［24］
Mueller M, Smith N, Ghanem B. A benchmark and simulator for uav tracking ［C］//European conference on computer vision. Springer, Cham, 2016: 445-461.
［25］
Guo D, Wang J, Cui Y, et al. SiamCAR: Siamese fully convolutional classification and regression for visual tracking ［C］//IEEE/CVF conference on computer vision and pattern recognition. 2020: 6269-6277.
［15］
Xu Y, Wang Z, Li Z, et al. Siamfc++: Towards robust and accurate visual tracking with target estimation guidelines ［C］//AAAI conference on Artificial Intelligence. 2020, 34(7): 12549-12556.
2023, Vol. 44 
