Adaptive foreground and shadow segmentation using hidden conditional random fields-中国期刊网

摘要 Videoobjectsegmentationisimportantforvideosurveillance,objecttracking,videoobjectrecognitionandvideoediting.Anadaptivevideosegmentationalgorithmbasedonhiddenconditionalrandomfields(HCRFs)isproposed,whichmodelsspatio-temporalconstraintsofvideosequence.Inordertoimprovethesegmentationquality,theweightsofspatio-temporalcon-straintsareadaptivelyupdatedbyon-linelearningforHCRFs.Shadowsarethefactorsaffectingsegmentationquality.Toseparateforegroundobjectsfromtheshadowstheycast,lineartransformforGaussiandistributionofthebackgroundisadoptedtomodeltheshadow.Theexperimentalresultsdemonstratedthattheerrorratioofouralgorithmisreducedby23%and19%respectively,comparedwiththeGaussianmixturemodel(GMM)andspatio-temporalMarkovrandomfields(MRFs).

1Wang Haochang Zhao Tiejun Li Sheng Yu Hao. A CONDITIONAL RANDOM FIELDS APPROACH TO BIOMEDICAL NAMED ENTITY RECOGNITION.物理电子学,2007-06.
2LIU Yi;LIU Xiao-hui. Testing conditional independence with data missing at random.基础数学,2018-03.
3Zhang Lili Zhang Yanning Wang Min Li Ying. ADAPTIVE RIVER SEGMENTATION IN SAR IMAGES.物理电子学,2009-04.
4Zhi-yuan Huang Chu-jin Li. Anisotropic Fractional Brownian Random Fields as White Noise Functionals.基础数学,2005-04.
5FU Ke-ang ZHANG Li-xin. Precise asymptotics in the law of the logarithm for random fields in Hilbert space.物理,2007-04.
6. Central Limit Theorems for Asymptotically Negatively Associated Random Fields.基础数学,2000-04.
7陈冬青. A CENTRAL LIMIT THEOREM FOR LATTICE GAUGE GIBBS RANDOM FIELDS.基础数学,1994-02.
8HESHUYUAN. UNIFORM CONVERGENCY FOR WEIGHTED PERIODOGRAM OF STATIONARY LINEAR RANDOM FIELDS.基础数学,1995-03.
9Zhen-long Chen;San-yang Liu. Hausdorff-type Measures of the Sample Path of Gaussian Random Fields.基础数学,2005-04.
10ZhangLixin WangXiuyun. CONVERGENCE RATES IN THE STRONG LAWS OF ASYMPTOTICALLY NEGATIVELY ASSOCIATED RANDOM FIELDS.基础数学,1999-04.

Adaptive foreground and shadow segmentation using hidden conditional random fields

来源期刊

相关推荐

同分类资源更多

相关关键词

Adaptive foreground and shadow segmentation using hidden conditional random fields

来源期刊

相关推荐

同分类资源 更多

相关关键词

同分类资源更多