Flow Mosaicking: Real-time Pedestrian Counting without Scene-specific Learning

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 机器人学研究室

	Flow Mosaicking: Real-time Pedestrian Counting without Scene-specific Learning
	Cong Y(丛杨); Gong HF(龚海峰); Zhu SC(朱松纯); Tang YD(唐延东)
	2009
会议名称	IEEE-Computer-Society Conference on Computer Vision and Pattern Recognition Workshops
会议日期	June 20-25, 2009
会议地点	Miami Beach, FL
页码	1093-1100
中文摘要	In this paper, we present a novel algorithm based on flow velocity field estimation to count the number of pedestrians across a detection line or inside a specified region. We regard pedestrians across the line as fluid flow, and design a novel model to estimate the flow velocity field. By integrating over time, the dynamic mosaics are constructed to count the number of pixels and edges passed through the line. Consequentially, the number of pedestrians can be estimated by quadratic regression, with the number of weighted pixels and edges as input. The regressors are learned off line from several camera tilt angles, and have taken the calibration information into account. We use tilt-angle-specific learning to ensure direct deployment and avoid overfitting while the commonly used scene-specific learning scheme needs on-site annotation and always trends to overfitting. Experiments on a variety of videos verified that the proposed method can give accurate estimation under different camera setup in real-time.
收录类别	EI ; CPCI（ISTP）
产权排序	1
会议主办者	IEEE Comp Soc
会议录	CVPR: 2009 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, VOLS 1-4
会议录出版者	IEEE
会议录出版地	NEW YORK
语种	英语
ISSN号	1063-6919
ISBN号	978-1-4244-3992-8
WOS记录号	WOS:000279038000140
内容类型	会议论文
源URL	[http://ir.sia.cn/handle/173321/8717]
专题	沈阳自动化研究所_机器人学研究室
推荐引用方式 GB/T 7714	Cong Y,Gong HF,Zhu SC,et al. Flow Mosaicking: Real-time Pedestrian Counting without Scene-specific Learning[C]. 见:IEEE-Computer-Society Conference on Computer Vision and Pattern Recognition Workshops. Miami Beach, FL. June 20-25, 2009.