标题:A two-step cross-modal hashing by exploiting label correlations and preserving similarity in both steps
作者:Chen, Zhen-Duo ;Luo, Xin ;Wang, Yongxin ;Nie, Liqiang ;Li, Hui-Qiong ;Xu, Xin-Shun
通讯作者:Xu, XinShun
作者机构:[Chen, Zhen-Duo ;Luo, Xin ;Wang, Yongxin ;Li, Hui-Qiong ;Xu, Xin-Shun ] School of Software, Shandong University, Jinan, China;[Nie, Liqiang ] School o 更多
会议名称:27th ACM International Conference on Multimedia, MM 2019
会议日期:21 October 2019 through 25 October 2019
来源:MM 2019 - Proceedings of the 27th ACM International Conference on Multimedia
出版年:2019
页码:1694-1702
DOI:10.1145/3343031.3350862
关键词:Cross-Modal Retrieval; Discrete Optimization; Label Correlations; Scalability; Similarity Preserving; Two-Step Hashing
摘要:In this paper, we present a novel Two-stEp Cross-modal Hashing method, TECH for short, for cross-modal retrieval tasks. As a two-step method, it first learns hash codes based on semantic labels, while preserving the similarity in the original space and exploiting the label correlations in the label space. In the light of this, it is able to make better use of label information and generate better binary codes. In addition, different from other two-step methods that mainly focus on the hash codes learning, TECH adopts a new hash function learning strategy in the second step, which also preserves the similarity in the original space. Moreover, with the help of well designed objective function and optimization scheme, it is able to generate hash codes discretely and scalable for large scale data. To the best of our knowledge, it is the first cross-modal hashing method exploiting label correlations, and also the first two-step hashing model preserving the similarity while leaning hash function. Extensive experiments demonstrate that the proposed approach outperforms some state-of-the-art cross-modal hashing methods. © 2019 Association for Computing Machinery.
收录类别:EI;SCOPUS
资源类型:会议论文;期刊论文
原文链接:https://www.scopus.com/inward/record.uri?eid=2-s2.0-85074863291&doi=10.1145%2f3343031.3350862&partnerID=40&md5=1c87d881d2269a91a272cb16f518a1cf
TOP