Learning to Diffuse: A New Perspective to Design PDEs for Visual   Analysis

doi:10.1109/TPAMI.2016.2522415

CORC > 北京大学 > 信息科学技术学院

	Learning to Diffuse: A New Perspective to Design PDEs for Visual Analysis
	Liu, Risheng ; Zhong, Guangyu ; Cao, Junjie ; Lin, Zhouchen ; Shan, Shiguang ; Luo, Zhongxuan
刊名	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
	2016
关键词	Visual diffusion PDE governed combinatorial optimization submodularity saliency detection object tracking SALIENT REGION DETECTION IMAGE SEGMENTATION TRACKING MODEL
DOI	10.1109/TPAMI.2016.2522415
英文摘要	Partial differential equations (PDEs) have been used to formulate image processing for several decades. Generally, a PDE system consists of two components: the governing equation and the boundary condition. In most previous work, both of them are generally designed by people using mathematical skills. However, in real world visual analysis tasks, such predefined and fixed-form PDEs may not be able to describe the complex structure of the visual data. More importantly, it is hard to incorporate the labeling information and the discriminative distribution priors into these PDEs. To address above issues, we propose a new PDE framework, named learning to diffuse (LTD), to adaptively design the governing equation and the boundary condition of a diffusion PDE system for various vision tasks on different types of visual data. To our best knowledge, the problems considered in this paper (i.e., saliency detection and object tracking) have never been addressed by PDE models before. Experimental results on various challenging benchmark databases show the superiority of LTD against existing state-of-the-art methods for all the tested visual analysis tasks.; National Natural Science Foundation of China (NSFC) [61300086, 61432003]; Fundamental Research Funds for the Central Universities [DUT15QY15]; Hong Kong Scholar Program [XJ2015008]; China Scholarship Council; NSFC [61363048, 61272341, 61231002, 61222211]; National Basic Research Program of China (973 Program) [2015CB352502]; Microsoft Research Asia Collaborative Research Program; SCI(E); ARTICLE; rsliu@dlut.edu.cn; guangyuzhonghikari@gmail.com; jjcao1231@gmail.com; zlin@pku.edu.cn; sgshan@ict.ac.cn; zxluo@dlut.edu.cn; 12; 2457-2471; 38
语种	英语
内容类型	期刊论文
源URL	[http://ir.pku.edu.cn/handle/20.500.11897/458333]
专题	信息科学技术学院
推荐引用方式 GB/T 7714	Liu, Risheng,Zhong, Guangyu,Cao, Junjie,et al. Learning to Diffuse: A New Perspective to Design PDEs for Visual Analysis[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2016.
APA	Liu, Risheng,Zhong, Guangyu,Cao, Junjie,Lin, Zhouchen,Shan, Shiguang,&Luo, Zhongxuan.(2016).Learning to Diffuse: A New Perspective to Design PDEs for Visual Analysis.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE.
MLA	Liu, Risheng,et al."Learning to Diffuse: A New Perspective to Design PDEs for Visual Analysis".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2016).