Beyond Group: Multiple Person Tracking via Minimal Topology-Energy-Variation

doi:10.1109/TIP.2017.2708901

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 视频内容安全

	Beyond Group: Multiple Person Tracking via Minimal Topology-Energy-Variation
	Gao, Shan 1; Ye, Qixiang 2; Xing, Junliang3 ; Kuijper, Arjan 4; Han, Zhenjun 2; Jiao, Jianbin 2; Ji, Xiangyang 1
发表期刊	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2017-12-01
卷号	26 期号:12 页码:5575-5589
文章类型	Article
摘要	Tracking multiple persons is a challenging task when persons move in groups and occlude each other. Existing group-based methods have extensively investigated how to make group division more accurately in a tracking-by-detection framework; however, few of them quantify the group dynamics from the perspective of targets' spatial topology or consider the group in a dynamic view. Inspired by the sociological properties of pedestrians, we propose a novel socio-topology model with a topology-energy function to factor the group dynamics of moving persons and groups. In this model, minimizing the topology-energy-variance in a two-level energy form is expected to produce smooth topology transitions, stable group tracking, and accurate target association. To search for the strong minimum in energy variation, we design the discrete group-tracklet jump moves embedded in the gradient descent method, which ensures that the moves reduce the energy variation of group and trajectory alternately in the varying topology dimension. Experimental results on both RGB and RGB-D data sets show the superiority of our proposed model for multiple person tracking in crowd scenes.
关键词	Multiple Person Tracking Group Tracking Rgb-d Data Topology
WOS标题词	Science & Technology ; Technology
DOI	10.1109/TIP.2017.2708901
关键词[WOS]	MULTITARGET TRACKING ; MULTIOBJECT TRACKING ; MULTIPERSON TRACKING ; PEDESTRIAN DETECTION ; DATA ASSOCIATION ; CRF MODEL ; MOTION
收录类别	SCI
语种	英语
项目资助者	National Natural Science Foundation of China(61325003 ; China Post-doctoral Science Foundation(2016M601028) ; 61620106005 ; 61601466 ; 61672519)
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000409526000002
引用统计	被引频次：8[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/20031
专题	多模态人工智能系统全国重点实验室_视频内容安全
作者单位	1.Tsinghua Univ, Dept Automat, Beijing 100084, Peoples R China 2.Univ Chinese Acad Sci, Sch Elect Elect & Commun Engn, Beijing 100049, Peoples R China 3.Chinese Acad Sci CASIA, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 4.Technol Univ Darmstadt, Fraunhofer Inst Comp Graph Res, D-64283 Darmstadt, Germany
推荐引用方式 GB/T 7714	Gao, Shan,Ye, Qixiang,Xing, Junliang,et al. Beyond Group: Multiple Person Tracking via Minimal Topology-Energy-Variation[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2017,26(12):5575-5589.
APA	Gao, Shan.,Ye, Qixiang.,Xing, Junliang.,Kuijper, Arjan.,Han, Zhenjun.,...&Ji, Xiangyang.(2017).Beyond Group: Multiple Person Tracking via Minimal Topology-Energy-Variation.IEEE TRANSACTIONS ON IMAGE PROCESSING,26(12),5575-5589.
MLA	Gao, Shan,et al."Beyond Group: Multiple Person Tracking via Minimal Topology-Energy-Variation".IEEE TRANSACTIONS ON IMAGE PROCESSING 26.12(2017):5575-5589.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
TIP17BeyondGroup.pdf（5475KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载