Variational Bayesian Inference for Finite Inverted Dirichlet Mixture Model and Its Application to Object Detection  

作　　者：LAI Yuping PING Yuan HE Wenda WANG Baocheng WANG Jingzhong ZHANG Xiufeng 

机构地区：[1]Department of Information Security, North China University of Technology, Beijing 100144, China [2]School of Information Engineering, Xuchang University, Xuchang 461000, China [3]National Research Center for Rehabilitation Technical Aids, Beijing 100176, China

出　　处：《Chinese Journal of Electronics》2018年第3期603-610,共8页电子学报（英文版）

基　　金：the National Natural Science Foundation of China(No.51335004,No.61363085,No.61303232);the Project of Action Plan Powerful School with Talents in North China University of Technology(No.XN018022);the Project of Science and Technology Innovation Service Capacity Building Project(No.PXM2017-014212-000002);the Program for Science&Technology Innovation Talents in Universities of Henan Province(No.18HASTIT022);the Foundation of Henan Educational Committee(No.16A520025,No.18A520047)

摘　　要：As a variant of Finite mixture model(FMM), finite Inverted Dirichlet mixture model(IDMM)can not avoid the conventional challenges, such as how to select the appropriate number of mixture components based on the observed data. Towards easing these issues,we propose a variational inference framework for learning IDMM which has been proved to be an efficient tool for modeling vectors with positive elements. Compared with the conventional Expectation maximization(EM) algorithm commonly used for learning FMM, the proposed approach prevents over-fitting well. Furthermore, it is able to do automatic determination of the number of mixture components and parameters estimation, simultaneously.Experimental results on both synthetic and real data of object detection confirm significant improvements on flexibility and efficiency being achieved.As a variant of Finite mixture model（FMM）, finite Inverted Dirichlet mixture model（IDMM）can not avoid the conventional challenges, such as how to select the appropriate number of mixture components based on the observed data. Towards easing these issues,we propose a variational inference framework for learning IDMM which has been proved to be an efficient tool for modeling vectors with positive elements. Compared with the conventional Expectation maximization（EM） algorithm commonly used for learning FMM, the proposed approach prevents over-fitting well. Furthermore, it is able to do automatic determination of the number of mixture components and parameters estimation, simultaneously.Experimental results on both synthetic and real data of object detection confirm significant improvements on flexibility and efficiency being achieved.

关 键 词：Bayesian estimation Mixture models Inverted Dirichlet distribution Variational inference Object detection 

分 类 号：O212.8[理学—概率论与数理统计]