Learning a Probabilistic Topology Discovering Model for Scene Categorization

CORC > 西安光学精密机械研究所 > 中国科学院西安光学精密机械研究所 > 光学影像学习与分析中心

	Learning a Probabilistic Topology Discovering Model for Scene Categorization
	Zhang, Luming 1; Ji, Rongrong 2; Xia, Yingjie 3; Zhang, Ying 1; Li, Xuelong4
刊名	ieee transactions on neural networks and learning systems
	2015-08-01
卷号	26 期号:8 页码:1622-1634
关键词	Boosting discrimination learning probabilistic model redundancy topology
英文摘要	a recent advance in scene categorization prefers a topological based modeling to capture the existence and relationships among different scene components. to that effect, local features are typically used to handle photographing variances such as occlusions and clutters. however, in many cases, the local features alone cannot well capture the scene semantics since they are extracted from tiny regions (e.g., 4 x 4 patches) within an image. in this paper, we mine a discriminative topology and a low-redundant topology from the local descriptors under a probabilistic perspective, which are further integrated into a boosting framework for scene categorization. in particular, by decomposing a scene image into basic components, a graphlet model is used to describe their spatial interactions. accordingly, scene categorization is formulated as an intergraphlet matching problem. the above procedure is further accelerated by introducing a probabilistic based representative topology selection scheme that makes the pairwise graphlet comparison trackable despite their exponentially increasing volumes. the selected graphlets are highly discriminative and independent, characterizing the topological characteristics of scene images. a weak learner is subsequently trained for each topology, which are boosted together to jointly describe the scene image. in our experiment, the visualized graphlets demonstrate that the mined topological patterns are representative to scene categories, and our proposed method beats state-of-the-art models on five popular scene data sets.
WOS标题词	science & technology ; technology
类目[WOS]	computer science, artificial intelligence ; computer science, hardware & architecture ; computer science, theory & methods ; engineering, electrical & electronic
研究领域[WOS]	computer science ; engineering
关键词[WOS]	object recognition ; image features ; aerial image ; annotation ; histograms ; algorithm
收录类别	SCI ; EI
语种	英语
WOS记录号	WOS:000358224200004
公开日期	2015-09-15
内容类型	期刊论文
源URL	[http://ir.opt.ac.cn/handle/181661/25258]
专题	西安光学精密机械研究所_光学影像学习与分析中心
作者单位	1.Natl Univ Singapore, Sch Comp, Singapore 119077, Singapore 2.Xiamen Univ, Sch Informat Sci & Engn, Dept Cognit Sci, Xiamen 361000, Peoples R China 3.Zhejiang Univ, Dept Comp Sci, Hangzhou 310027, Zhejiang, Peoples R China 4.Chinese Acad Sci, State Key Lab Transient Opt & Photon, Ctr Opt Imagery Anal & Learning, Xian Inst Opt & Precis Mech, Xian 710119, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Luming,Ji, Rongrong,Xia, Yingjie,et al. Learning a Probabilistic Topology Discovering Model for Scene Categorization[J]. ieee transactions on neural networks and learning systems,2015,26(8):1622-1634.
APA	Zhang, Luming,Ji, Rongrong,Xia, Yingjie,Zhang, Ying,&Li, Xuelong.(2015).Learning a Probabilistic Topology Discovering Model for Scene Categorization.ieee transactions on neural networks and learning systems,26(8),1622-1634.
MLA	Zhang, Luming,et al."Learning a Probabilistic Topology Discovering Model for Scene Categorization".ieee transactions on neural networks and learning systems 26.8(2015):1622-1634.