基于改进Tracktor的行人多目标跟踪算法

doi:10.3778/j.issn.1002-8331.2212-0096

摘要/Abstract

摘要： 在多目标视频跟踪中，针对受交互遮挡等影响导致检测偏差从而致使目标身份丢失的问题，提出一种基于改进Tracktor的行人多目标跟踪算法DUTracktor。在检测框回归中设计一个动态更新模块，利用孪生网络对建议框进一步检测定位；利用时序信息增强模块更新当前帧更适合的模板，建立全局上下文关系；并通过像素相关进行特征融合，从而增强目标边缘信息和尺度信息；利用相机运动补偿和融合相似矩阵构建二级关联跟踪机制，建立检测框和轨迹更强大的关联性，提高目标跟踪的鲁棒性。在公开的MOT16数据集上进行实验测试，并与当前主流算法相比，该算法跟踪精度表现较优，具有良好的鲁棒性，FPS稳定在24帧。

关键词: 计算机视觉, 多目标跟踪, Tracktor, 孪生网络

Abstract: In multi-target video tracking, for the problem of detection bias caused by interaction occlusion and other influences, thus resulting in target identity loss, an improved Tracktor-based pedestrian multi-target tracking algorithm is proposed. Firstly, a dynamic update module is designed in the detection frame regression to further detect and locate the proposed frame by using twin networks. Then, the temporal information enhancement module is used to update a more suitable template for current frame and establish global contextual relationships. And feature fusion is performed through pixel correlation, thus enhancing target edge information and scale information. Finally, camera motion compensation and fusion similarity matrix are adopted to construct a secondary correlation tracking mechanism to establish stronger correlation between detection frame and trajectory, and improve the robustness of target tracking. Experimental tests are conducted on public available MOT16 dataset, in comparasion with current mainstream algorithms, the tracking accuracy performance of the proposed algorithm is better,which has good robustness with a stable FPS of 24 frames.

Key words: computer vision, multi-target tracking, Tracktor, siamese network

谌海云, 黄忠义, 王海川, 余鸿皓. 基于改进Tracktor的行人多目标跟踪算法[J]. 计算机工程与应用, 2024, 60(8): 242-249.

SHEN Haiyun, HUANG Zhongyi, WANG Haichuan, YU Honghao. Improved Tracktor-Based Pedestrian Multi-Objective Tracking Algorithm[J]. Computer Engineering and Applications, 2024, 60(8): 242-249.

参考文献

[1] 柳恩涵, 张锐, 赵硕, 等. 一种基于视频预测的红外行人目标跟踪方法[J].哈尔滨工业大学学报, 2020, 52(10): 192-200.
LIU E H, ZHANG R, ZHAO S, et al. Infrared pedestrian target tracking method based on video prediction[J]. Journal of Harbin Institute of Technology, 2020, 52(10): 192-200.
[2] YU E, LI Z, HAN S, et al. RelationTrack: relation-aware multiple object tracking with decoupled representation[J]. IEEE Transactions on Multimedia, 2023, 25: 2686-2697.
[3] TAN C, LI C, HE D, et al. Towards real-time tracking and counting of seedlings with a one-stage detector and optical flow[J]. Computers and Electronics in Agriculture, 2022, 193: 106683.
[4] ZHOU X, YIN T, KOLTUN V, et al. Global tracking transformers[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 8771-8780.
[5] BRASó G, LEAL-TAIXé L. Learning a neural solver for multiple object tracking[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 6247-6257.
[6] 陈璠, 彭力.多层级重叠条纹特征融合的行人重识别[J].计算机科学与探索, 2021, 15(9): 1753-1761.
CHEN F, PENG L. Person re-identification based on multi-level feature fusion with overlapping stripes[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(9): 1753-1761.
[7] 李杰.结合注意力和纹理特征增强的行人再识别[J].计算机科学与探索, 2022, 16(3): 661-668.
LI J. Attention and texture feature enhancement for person re-identification[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(3): 661-668.
[8] TANG S, ANDRILUKA M, ANDRES B, et al. Multiple people tracking by lifted multicut and person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 3539-3548.
[9] SRIDHAR V H, ROCHE D G, GINGINS S. Tracktor: image-based automated tracking of animal movement and behaviour[J]. Methods in Ecology and Evolution, 2019, 10(6): 815-820.
[10] WANG L, XU L, KIM M Y, et al. Online multiple object tracking via flow and convolutional features[C]//Proceedings of the 2017 IEEE International Conference on Image Processing, 2017: 3630-3634.
[11] FERNANDO T, DENMAN S, SRIDHARAN S, et al. Tracking by prediction: a deep generative model for mutli-person localisation and tracking[C]//Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision, 2018: 1122-1132.
[12] ZHAO D, FU H, XIAO L, et al. Multi-object tracking with correlation filter for autonomous vehicle[J]. Sensors, 2018, 18(7): 2004.
[13] BERTINETTO L, VALMADRE J, HENRIQUES J F, et al. Fully-convolutional siamese networks for object tracking[C]//Proceedings of the European Conference on Computer Vision, 2016: 850-865.
[14] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[15] JACOB D, CHANG M, LEE K. Bert: pre-training of deep bidirectional transformers for language understanding[C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2019: 4171-4186.
[16] LIN H, CHENG X, WU X, et al. Cat: cross attention in vision transformer[C]//Proceedings of the 2022 IEEE International Conference on Multimedia and Expo, 2022: 1-6.
[17] REZATOFIGHI H, TSOI N, GWAK J Y, et al. Generalized intersection over union: a metric and a loss for bounding box regression[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 658-666.
[18] AHARON N, ORFAIG R, BOBROVSKY B Z. BoT-SORT: robust associations multi-pedestrian tracking[J]. arXiv:2206.14651, 2022.
[19] CHU P, FAN H, TAN C C, et al. Online multi-object tracking with instance-aware tracker and dynamic model refreshment[C]//Proceedings of the 2019 IEEE Winter Conference on Applications of Computer Vision, 2019: 161-170.
[20] XU Y, OSEP A, BAN Y, et al. How to train your deep multi-object tracker[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 6787-6796.
[21] WANG Z, ZHENG L, LIU Y, et al. Towards real-time multi-object tracking[C]//Proceedings of the European Conference on Computer Vision, 2020: 107-122.
[22] LIU Q, CHEN D, CHU Q, et al. Online multi-object tracking with unsupervised re-identification learning and occlusion estimation[J]. arXiv:2201.01297, 2022.
[23] ZHANG Y, WANG C, WANG X, et al. FairMOT: on the fairness of detection and re-identification in multiple object tracking[J]. International Journal of Computer Vision, 2021, 129(11): 3069-3087.