View-invariant Action Recognition Using Cross Ratios Across Frames

CASIA OpenIR > 模式识别实验室

	View-invariant Action Recognition Using Cross Ratios Across Frames
	Yeyin Zhang; Kaiqi Huang; Yongzhen Huang; Tieniu Tan
	2009
会议名称	International Conference on Image Processing
会议录名称	IEEE International Conference on Image Processing, 2009
页码	3549--3552
会议日期	2009
会议地点	Cairo, Egypt
摘要	We present a new method of computing invariants in videos captured from different views to achieve view-invariant action recognition. To avoid the constraints of collinearity or coplanarity of image points for constructing invariants, we consider several neighboring frames to compute cross ratios, namely cross ratios across frames (CRAF), as our invariant representation of action. For every five points sampled with different intervals from the trajectories of action, we construct a pair of cross ratios (CRs). Afterwards, we transform the CRs to histograms as the feature vectors for classification. Experimental results demonstrate that the proposed method outperforms the state-of-the-art methods in effectiveness and stability.
关键词	Computational Geometry gesture Recognition image Classification
语种	英语
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/12707
专题	模式识别实验室
通讯作者	Kaiqi Huang
作者单位	中国科学院自动化研究所
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Yeyin Zhang,Kaiqi Huang,Yongzhen Huang,et al. View-invariant Action Recognition Using Cross Ratios Across Frames[C],2009:3549--3552.