SCALE-ADAPTIVE CNN BASED CROWD COUNTING AND DYNAMIC SUPERVISION

doi:10.1109/ICMEW.2019.00076

	SCALE-ADAPTIVE CNN BASED CROWD COUNTING AND DYNAMIC SUPERVISION
	Li, Zhengxin1 ; Li, Jing1 ; Xie, Ling1 ; Liu, Jianli 2
	2019
会议录名称	2019 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA & EXPO WORKSHOPS (ICMEW)
页码	408-413
发表状态	已发表
DOI	10.1109/ICMEW.2019.00076
摘要	A major problem in crowd counting is the varying size of human heads in an image due to the changeable camera viewpoint, perspective and imaging resolution. To resolve this problem, we propose scale-adaptive CNN-based framework (SA-CNN) which implicitly outputs features at multiple scales and adaptively estimates density maps from the fusion of these features. To overcome the fact that head size is not annotated in all the public crowd datasets, we design a dynamic supervision strategy to reach the goal. Experiment results show that our method achieve state-of-the-art performance while demonstrating significant robustness to hyper-parameters even the choice of backbone.
关键词	Crowd counting scale-adaptive dynamic supervision
会议地点	Shanghai, China
会议日期	8-12 July 2019
URL	查看原文
收录类别	CPCI ; CPCI-S ; EI
语种	英语
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Information Systems ; Engineering, Electrical & Electronic
WOS记录号	WOS:000490430500069
出版者	IEEE
原始文献类型	Proceedings Paper
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/80392
专题	生命科学与技术学院_博士生信息科学与技术学院_PI研究组_高盛华组物质科学与技术学院_本科生
通讯作者	Li, Zhengxin
作者单位	1.ShanghaiTech Univ, Shanghai, Peoples R China 2.Jiangnan Univ, Wuxi, Jiangsu, Peoples R China
第一作者单位	上海科技大学
通讯作者单位	上海科技大学
第一作者的第一单位	上海科技大学
推荐引用方式 GB/T 7714	Li, Zhengxin,Li, Jing,Xie, Ling,et al. SCALE-ADAPTIVE CNN BASED CROWD COUNTING AND DYNAMIC SUPERVISION[C]:IEEE,2019:408-413.