基于双模型竞争机制的目标跟踪算法

doi:10.12263/DZXB.20221375

摘要/Abstract

摘要：

为解决背景感知相关滤波器存在的特征表达能力不足和模型漂移问题，本文提出了一种基于双模型竞争机制的目标跟踪算法.一方面，本文基于颜色和梯度信息设计了一种简单高效的特征描述子，以实现更鲁棒的目标表观建模.另一方面，本文分别构建初始模型和变化模型作用于目标搜索区域，并根据两者的跟踪响应图置信度来决定跟踪结果.跟踪过程中，随着双模型主导地位不断地动态切换，变化模型也被赋予了可逆向学习的能力，从而达到缓解模型漂移的效果.实验结果表明，相比于基准算法，本文算法在OTB2015、TinyTLP和UAV20L三个数据集的跟踪精度分别提升5.0%、1.3%和4.1%，跟踪成功率分别提升3.8%、2.8%和1.7%，且在对不同跟踪场景实现稳定跟踪的同时能够保持25.5 fps的实时跟踪速度.

Abstract:

To solve the problems of insufficient feature expression ability and model drift in the background-aware correlation filters, this paper proposes an object tracking algorithm based on a dual-model competition mechanism.On the one hand, a simple and efficient feature descriptor that integrates color and gradient information is designed to achieve more robust target appearance modeling.On the other hand, we construct two filter models to describe the object's initial appearance and its variations, and then apply them to the target searching area respectively.The tracking results are determined by the confidence of the tracking response maps corresponding to these two models.During the tracking process, with the dynamic switching of the dominant position of the two models, the filter model for adapting to object variations is also endowed with the ability of reversible learning to alleviate the model drift.The experimental results show that, compared with the baseline tracker, the tracking precision of the proposed algorithm on OTB2015, TinyTLP and UAV20L datasets is improved by 5.0%, 1.3% and 4.1%, and the tracking success rate is improved by 3.8%,2.8% and 1.7%.The proposed algorithm can also achieve stable tracking performance for different tracking scenarios while maintaining a running speed of 25.5 fps.

Key words: object tracking, correlation filter, dual-model competition mechanism, feature descriptor, tracking confidence, model drift

中图分类号:

TP391.4

林彬, 王华通, 封全喜. 基于双模型竞争机制的目标跟踪算法[J]. 电子学报, DOI: 10.12263/DZXB.20221375.

Bin LIN, Hua-tong WANG, Quan-xi FENG . Object Tracking Algorithm Based on Dual-Model Competition Mechanism[J]. Acta Electronica Sinica, DOI: 10.12263/DZXB.20221375.

图/表 8

参考文献 22

1	韩瑞泽, 冯伟, 郭青, 等. 视频单目标跟踪研究进展综述[J]. 计算机学报, 2022, 45(9): 1877-1907.
	HAN R Z, FENG W, GUO Q, et al. Single object tracking research: A survey[J]. Chinese Journal of Computers, 2022, 45(9): 1877-1907. (in Chinese)
2	BOLME D S, BEVERIDGE J R, DRAPER B A, et al. Visual object tracking using adaptive correlation filters[C]//2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2010: 2544-2550.
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, 2015, 37(3): 583-596.
4	仇祝令, 查宇飞, 吴敏, 等. 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 Z L, ZHA Y F, WU M, et al. Learning attentional regularized correlation filter for visual tracking[J]. Acta Electronica Sinica, 2020, 48(9): 1762-1768. (in Chinese)
5	GALOOGAHI H K, FAGG A, LUCEY S. Learning background-aware correlation filters for visual tracking[C]//2017 IEEE International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2017: 1144-1152
6	DANELLJAN M, KHAN F S, FELSBERG M, et al. Adaptive color attributes for real-time visual tracking[C]//2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2014: 1090-1097.
7	LI Y, ZHU J K. A Scale Adaptive Kernel Correlation Filter Tracker with Feature Integration[C]//European Conference on Computer Vision Workshops. Cham: Springer, 2015: 254-265.
8	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 (CVPR). Piscataway: IEEE, 2016: 1401-1409.
9	MA C, HUANG J B, YANG X K, et al. Hierarchical convolutional features for visual tracking[C]//2015 IEEE International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2016: 3074-3082.
10	蒲磊, 冯新喜, 侯志强, 等. 基于二阶池化网络的鲁棒视觉跟踪算法[J]. 电子学报, 2020, 48(8): 1472-1478.
	PU L, FENG X X, HOU Z Q, et al. Robust visual tracking based on second order pooling network[J]. Acta Electronica Sinica, 2020, 48(8): 1472-1478. (in Chinese)
11	蒲磊, 冯新喜, 侯志强, 等. 基于深度空间正则化的相关滤波跟踪算法[J]. 电子学报, 2020, 48(10): 2025-2032.
	PU L, FENG X X, HOU Z Q, et al. Correlation filter tracking based on deep spatial regularization[J]. Acta Electronica Sinica, 2020, 48(10): 2025-2032. (in Chinese)
12	XU T Y, FENG Z H, WU X J, et al. Adaptive channel selection for robust visual object tracking with discriminative correlation filters[J]. International Journal of Computer Vision, 2021, 129(5): 1359-1375.
13	WANG N, ZHOU W G, LI H Q. Reliable re-detection for long-term tracking[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2019, 29(3): 730-743.
14	WANG M M, LIU Y, HUANG Z Y. Large margin object tracking with circulant feature maps[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway: IEEE, 2017: 4800-4808.
15	WU Y, LIM J, YANG M H. Object tracking benchmark[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(9): 1834-1848.
16	MOUDGIL A, GANDHI V. Long-Term Visual Object Tracking Benchmark[C]//Asian Conference on Computer Vision. Cham: Springer, 2019: 629-645.
17	MUELLER M, SMITH N, GHANEM B. A Benchmark and Simulator for UAV Tracking[C]//European Conference on Computer Vision. Cham: Springer, 2016: 445-461.
18	DANELLJAN M, HÄGER G, KHAN F S, et al. Discriminative scale space tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(8): 1561-1575.
19	ZHU G B, WANG J Q, WU Y, et al. Collaborative correlation tracking[C]//Proceedings of the British Machine Vision Conference (BMVC). Durham: BMVA, 2015: 184.1-184.12.
20	LI F, TIAN C, ZUO W M, et al. Learning spatial-temporal regularized correlation filters for visual tracking[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2018: 4904-4913.
21	FU C H, XU J T, LIN F L, et al. Object saliency-aware dual regularized correlation filter for real-time aerial tracking[J]. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(12): 8940-8951.
22	HUANG Z Y, FU C H, LI Y M, et al. Learning aberrance repressed correlation filters for real-time UAV tracking[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Piscataway: IEEE, 2020: 2891-2900.

算法	IV	SV	OCC	DEF	MB	FM	IPR	OPR	OV	BC	LR
SAMF	0.728	0.717	0.737	0.688	0.692	0.669	0.718	0.718	0.747	0.704	0.709
STAPLE	0.782	0.731	0.728	0.751	0.719	0.709	0.767	0.739	0.668	0.749	0.591
ACS-DCF	0.802	0.849	0.832	0.790	0.844	0.800	0.818	0.848	0.790	0.827	0.770
RRDT	0.818	0.785	0.769	0.800	0.804	0.750	0.775	0.794	0.728	0.822	0.605
FDSST	0.751	0.670	0.636	0.610	0.711	0.698	0.734	0.668	0.577	0.780	0.617
CCT	0.742	0.695	0.681	0.657	0.666	0.668	0.710	0.702	0.591	0.749	0.650
STRCF	0.841	0.844	0.814	0.844	0.837	0.793	0.812	0.848	0.766	0.872	0.756
DRCF	0.744	0.704	0.647	0.765	0.729	0.742	0.703	0.701	0.597	0.735	0.599
ARCF	0.766	0.775	0.740	0.769	0.772	0.763	0.785	0.770	0.671	0.760	0.714
BACF	0.820	0.771	0.740	0.802	0.774	0.781	0.784	0.785	0.714	0.858	0.711
Ours	0.884	0.857	0.833	0.826	0.859	0.827	0.843	0.859	0.762	0.924	0.777

算法	IV	SV	OCC	DEF	MB	FM	IPR	OPR	OV	BC	LR
SAMF	0.728	0.717	0.737	0.688	0.692	0.669	0.718	0.718	0.747	0.704	0.709
STAPLE	0.782	0.731	0.728	0.751	0.719	0.709	0.767	0.739	0.668	0.749	0.591
ACS-DCF	0.802	0.849	0.832	0.790	0.844	0.800	0.818	0.848	0.790	0.827	0.770
RRDT	0.818	0.785	0.769	0.800	0.804	0.750	0.775	0.794	0.728	0.822	0.605
FDSST	0.751	0.670	0.636	0.610	0.711	0.698	0.734	0.668	0.577	0.780	0.617
CCT	0.742	0.695	0.681	0.657	0.666	0.668	0.710	0.702	0.591	0.749	0.650
STRCF	0.841	0.844	0.814	0.844	0.837	0.793	0.812	0.848	0.766	0.872	0.756
DRCF	0.744	0.704	0.647	0.765	0.729	0.742	0.703	0.701	0.597	0.735	0.599
ARCF	0.766	0.775	0.740	0.769	0.772	0.763	0.785	0.770	0.671	0.760	0.714
BACF	0.820	0.771	0.740	0.802	0.774	0.781	0.784	0.785	0.714	0.858	0.711
Ours	0.884	0.857	0.833	0.826	0.859	0.827	0.843	0.859	0.762	0.924	0.777

算法	精度	成功率
BACF	0.782	0.595
BACF + HOGCN	0.804	0.601
BACF + HSV-HOG	0.807	0.619
BACF + C-HSV-HOG	0.819	0.625

算法	精度	成功率
BACF	0.782	0.595
BACF + HOGCN	0.804	0.601
BACF + HSV-HOG	0.807	0.619
BACF + C-HSV-HOG	0.819	0.625

算法	精度	成功率	速度/fps
BACF	0.822	0.620	47.2
BACF + FI	0.852	0.643	26.6
BACF + FI + DM	0.872	0.658	25.5