Fusing Features and Local Features with Context-Aware Kernels for Action Recognition

doi:10.1007/s11263-015-0867-0

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

	Fusing Features and Local Features with Context-Aware Kernels for Action Recognition
	Yuan, Chunfeng1 ; Wu, Baoxin 1; Li, Xi 2; Hu, Weiming1 ; Maybank, Stephen 3; Wang, Fangshi 4
发表期刊	INTERNATIONAL JOURNAL OF COMPUTER VISION
	2016-06-01
卷号	118 期号:2 页码:151-171
文章类型	Article
摘要	The performance of action recognition in video sequences depends significantly on the representation of actions and the similarity measurement between the representations. In this paper, we combine two kinds of features extracted from the spatio-temporal interest points with context-aware kernels for action recognition. For the action representation, local cuboid features extracted around interest points are very popular using a Bag of Visual Words (BOVW) model. Such representations, however, ignore potentially valuable information about the global spatio-temporal distribution of interest points. We propose a new global feature to capture the detailed geometrical distribution of interest points. It is calculated by using the 3D transform which is defined as an extended 3D discrete Radon transform, followed by the application of a two-directional two-dimensional principal component analysis. For the similarity measurement, we model a video set as an optimized probabilistic hypergraph and propose a context-aware kernel to measure high order relationships among videos. The context-aware kernel is more robust to the noise and outliers in the data than the traditional context-free kernel which just considers the pairwise relationships between videos. The hyperedges of the hypergraph are constructed based on a learnt Mahalanobis distance metric. Any disturbing information from other classes is excluded from each hyperedge. Finally, a multiple kernel learning algorithm is designed by integrating the norm regularization into a linear SVM classifier to fuse the feature and the BOVW representation for action recognition. Experimental results on several datasets demonstrate the effectiveness of the proposed approach for action recognition.
关键词	Action Recognition Spatio-temporal Interest Points 3d r Transform Hypergraph Context-aware Kernel
WOS标题词	Science & Technology ; Technology
DOI	10.1007/s11263-015-0867-0
关键词[WOS]	TIME INTEREST POINTS ; FACE REPRESENTATION ; VISUAL RECOGNITION ; 2-DIMENSIONAL PCA ; CLASSIFICATION ; SVM
收录类别	SCI
语种	英语
项目资助者	973 basic research program of China(2014CB349303) ; Natural Science Foundation of China(61472421 ; CAS Center for Excellence in Brain Science and Intelligence Technology ; Guangdong Natural Science Foundation(S2012020011081) ; 61472420 ; 61303086 ; 61202327)
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000377477400004
引用统计	被引频次：7[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/173211/10838
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.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China 2.Zhejiang Univ, Coll Comp Sci & Technol, Hangzhou, Zhejiang, Peoples R China 3.Birkbeck Coll, Dept Comp Sci & Informat Syst, London, England 4.Beijing Jiaotong Univ, Sch Software Engn, Beijing, Peoples R China
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Yuan, Chunfeng,Wu, Baoxin,Li, Xi,et al. Fusing Features and Local Features with Context-Aware Kernels for Action Recognition[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2016,118(2):151-171.
APA	Yuan, Chunfeng,Wu, Baoxin,Li, Xi,Hu, Weiming,Maybank, Stephen,&Wang, Fangshi.(2016).Fusing Features and Local Features with Context-Aware Kernels for Action Recognition.INTERNATIONAL JOURNAL OF COMPUTER VISION,118(2),151-171.
MLA	Yuan, Chunfeng,et al."Fusing Features and Local Features with Context-Aware Kernels for Action Recognition".INTERNATIONAL JOURNAL OF COMPUTER VISION 118.2(2016):151-171.

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
IJCV 2015 Features a（3979KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览