Linear dependency modeling for classifier fusion and feature combination

doi:10.1109/TPAMI.2012.198

科研成果详情

题名	Linear dependency modeling for classifier fusion and feature combination
作者	Ma，Andy Jinhua 1; Yuen，Pong C.1,2; Lai，Jian Huang 3,4
发表日期	2013
发表期刊	IEEE Transactions on Pattern Analysis and Machine Intelligence
ISSN/eISSN	0162-8828
卷号	35 期号:5 页码:1135-1148
摘要	This paper addresses the independent assumption issue in fusion process. In the last decade, dependency modeling techniques were developed under a specific distribution of classifiers or by estimating the joint distribution of the posteriors. This paper proposes a new framework to model the dependency between features without any assumption on feature/classifier distribution, and overcomes the difficulty in estimating the high-dimensional joint density. In this paper, we prove that feature dependency can be modeled by a linear combination of the posterior probabilities under some mild assumptions. Based on the linear combination property, two methods, namely, Linear Classifier Dependency Modeling (LCDM) and Linear Feature Dependency Modeling (LFDM), are derived and developed for dependency modeling in classifier level and feature level, respectively. The optimal models for LCDM and LFDM are learned by maximizing the margin between the genuine and imposter posterior probabilities. Both synthetic data and real datasets are used for experiments. Experimental results show that LCDM and LFDM with dependency modeling outperform existing classifier level and feature level combination methods under nonnormal distributions and on four real databases, respectively. Comparing the classifier level and feature level fusion methods, LFDM gives the best performance. © 1979-2012 IEEE.
关键词	classifier level fusion feature dependency feature level fusion Linear dependency modeling multiple feature fusion
DOI	10.1109/TPAMI.2012.198
URL	查看来源
语种	英语English
Scopus入藏号	2-s2.0-84875433320
引用统计	被引频次：32[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	https://repository.uic.edu.cn/handle/39GCC9TT/6548
专题	北师香港浸会大学
作者单位	1.Department of Computer Science,Hong Kong Baptist University,Hong Kong,Hong Kong 2.BNU-HKBU United International College,Zhuhai,China 3.School of Information Science and Technology,Sun Yat-Sen University,Building 110,No. 135, Xin Gang Xi Road,Guangzhou 510257,China 4.Guangdong Province Key Laboratory of Information Security,Guangzhou 510006,China
推荐引用方式 GB/T 7714	Ma，Andy Jinhua,Yuen，Pong C.,Lai，Jian Huang. Linear dependency modeling for classifier fusion and feature combination[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(5): 1135-1148.
APA	Ma，Andy Jinhua, Yuen，Pong C., & Lai，Jian Huang. (2013). Linear dependency modeling for classifier fusion and feature combination. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(5), 1135-1148.
MLA	Ma，Andy Jinhua,et al."Linear dependency modeling for classifier fusion and feature combination". IEEE Transactions on Pattern Analysis and Machine Intelligence 35.5(2013): 1135-1148.