Image Tag Recommendation via Deep Cross-modal Correlation Mining

摘要：

　　In this paper,a novel image tag recommendation framework is developed by fusing the deep multimodal feature representation and cross-modal correlation mining,which enables the most appropriate and relevant tags to be presented on the image and facilitates more accurate image retrieval.Such an image tag recommendation pattern can be modeled as an inter-related correlation distribution over deep multimodal visual and semantic representations of images and tags,in which the most important is to create more effective cross-modal correlation and measure what degree they are related.Our experiments on a large number of public data have obtained very positive results.

关键词： Image Tag Recommendation Deep Multimodal Feature Representation Cross-modal Correlation Mining Deep Canonical Correlation Analysis

作者: Xingmeng Zhang Cheng Jin Yuejie Zhang Tao Zhang

作者单位: School of Computer Science Shanghai Key Laboratory of Intelligent Information Processing,Fudan Unive School of Information Management & Engineering,Shanghai University of Finance & Economics,Shanghai 2

会议类型: 国内会议

会议名称: 第十五届全国计算语言学学术会议(CCL2016)暨第四届基于自然标注大数据的自然语言处理国际学术研讨会(NLP-NABD-2016)

会议地点: 烟台

会议语种:英文

页码: 1-12

在线出版日期: 2016-10-14（万方平台首次上网日期，不代表论文的发表时间）