Top-down spatiotemporal saliency detection using spectral filtering

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 机器人应用与理论组

	Top-down spatiotemporal saliency detection using spectral filtering
	Li, Wanyi; Wang, Peng; Qiao, Hong
	2013
会议名称	5th International Conference on Digital Image Processing (ICDIP)
会议日期	APR 21-22, 2013
会议地点	Beijing, PEOPLES R CHINA
关键词	Visual attention Top-down spatiotemporal saliency Spectral filtering
通讯作者	Li, WY
英文摘要	A spectral filtering based method for top-down spatiotemporal saliency detection is proposed. The proposed method enables to favor the salient features of the target object needed to pop out. Here a feature vector representing the salient features of the target object is learned online within the first image in which it is detected or initialized manually. The proper scale of the Gaussian kernel forspectral filtering is selected automatically according to the size ratio of the whole image to the target object. Guided by the top-down information, a top-down, target-related saliency map can be built in subsequent images. This enables to focus on the most relevant salient region and can be extended to complicated computer vision tasks. Experiment results demonstrate the effectiveness of the proposed method.
会议录	FIFTH INTERNATIONAL CONFERENCE ON DIGITAL IMAGE PROCESSING (ICDIP 2013)
内容类型	会议论文
源URL	[http://ir.ia.ac.cn/handle/173211/12871]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_机器人应用与理论组
推荐引用方式 GB/T 7714	Li, Wanyi,Wang, Peng,Qiao, Hong. Top-down spatiotemporal saliency detection using spectral filtering[C]. 见:5th International Conference on Digital Image Processing (ICDIP). Beijing, PEOPLES R CHINA. APR 21-22, 2013.