Online Multiple Object Tracking with Cross-Task Synergy

doi:10.1109/CVPR46437.2021.00804

	Online Multiple Object Tracking with Cross-Task Synergy
	Guo, Song 1; Wang, Jingya1,2 ; Wang, Xinchao 3,4; Tao, Dacheng 1
	2021
会议录名称	2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
ISSN	1063-6919
发表状态	已发表
DOI	10.1109/CVPR46437.2021.00804
摘要	Modern online multiple object tracking (MOT) methods usually focus on two directions to improve tracking performance. One is to predict new positions in an incoming frame based on tracking information from previous frames, and the other is to enhance data association by generating more discriminative identity embeddings. Some works combined both directions within one framework but handled them as two individual tasks, thus gaining little mutual benefits. In this paper, we propose a novel unified model with synergy between position prediction and embedding association. The two tasks are linked by temporal-aware target attention and distractor attention, as well as identity-aware memory aggregation model. Specifically, the attention modules can make the prediction focus more on targets and less on distractors, therefore more reliable embeddings can be extracted accordingly for association. On the other hand, such reliable embeddings can boost identity-awareness through memory aggregation, hence strengthen attention modules and suppress drifts. In this way, the synergy between position prediction and embedding association is achieved, which leads to strong robustness to occlusions. Extensive experiments demonstrate the superiority of our proposed model over a wide range of existing methods on MOTChallenge benchmarks. Our code and models are publicly available at https://github.com/songguocod/TADAM.
会议名称	IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
出版地	10662 LOS VAQUEROS CIRCLE, PO BOX 3014, LOS ALAMITOS, CA 90720-1264 USA
会议地点	null,null,ELECTR NETWORK
会议日期	JUN 19-25, 2021
URL	查看原文
收录类别	CPCI-S ; EI ; CPCI
语种	英语
资助项目	Australian Research Council["FL-170100117","DP-180103424","IH-180100002","IC-190100031"]
WOS研究方向	Computer Science ; Imaging Science & Photographic Technology
WOS类目	Computer Science, Artificial Intelligence ; Imaging Science & Photographic Technology
WOS记录号	WOS:000739917308036
出版者	IEEE COMPUTER SOC
来源库	IEEE
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/153575
专题	信息科学与技术学院_PI研究组_汪婧雅组
通讯作者	Guo, Song; Wang, Jingya
作者单位	1.Univ Sydney, Sydney, NSW, Australia 2.ShanghaiTech Univ, Shanghai, Peoples R China 3.Natl Univ Singapore, Singapore, Singapore 4.Stevens Inst Technol, Hoboken, NJ 07030 USA
通讯作者单位	上海科技大学
推荐引用方式 GB/T 7714	Guo, Song,Wang, Jingya,Wang, Xinchao,et al. Online Multiple Object Tracking with Cross-Task Synergy[C]. 10662 LOS VAQUEROS CIRCLE, PO BOX 3014, LOS ALAMITOS, CA 90720-1264 USA:IEEE COMPUTER SOC,2021.