首页|Exploiting PLSA model and conditional random field for refining image annotation

Exploiting PLSA model and conditional random field for refining image annotation

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 维普
  • 万方数据
This paper presents a new method for refining image annotation by integrating probabilistic latent semantic analysis (PLSA) with conditional random field (CRF).First a PLSA model with asymmetric modalities is constructed to predict a candidate set of annotations with confidence scores,and then model semantic relationship among the candidate annotations by leveraging conditional random field.In CRF,the confidence scores generated by the PLSA model and the Flickr distance between pairwise candidate annotations are considered as local evidences and contextual potentials respectively.The novelty of our method mainly lies in two aspects:exploiting PLSA to predict a candidate setof annotations with confidence scores as well as CRF to further explore the semantic context among candidate annotations for precise image annotation.To demonstrate the effectiveness of the method proposed in thispaper,an experiment is conducted on the standard Corel dataset and its results are compared favorably with several state-of-the-art approaches.

automatic image annotationprobabilistic latent semantic analysis (PLSA)expectation-maximizationconditional random field(CRF)Flickr distanceimage retrieval

Tian Dongping

展开 >

Institute of Computer Software, Baoji University of Arts and Sciences, Baoji 721007, P.R.China

Institute of Computational Information Science, Baoji University of Arts and Sciences, Baoji 721007, P.R.China

National Basic Research Priorities ProgrammeNational High Technology Research and Development Programme of ChinaNatural Science Basic Research Plan in Shanxi Province of ChinaScience and Technology R&D Program of Baoji CityScience and Technology R&D Program of Baoji City

2013CB3295022012AA0110032014JQ2-60362030200132013R2-2

2015

10.3772/j.issn.1006-6748.2015.01.011

高技术通讯(英文版)
中国科学技术信息研究所（ISTIC）

高技术通讯(英文版)

EI
影响因子：0.058
ISSN：1006-6748
年,卷(期)：2015.21(1)
  • 1
  • 26