标题:MODALITY-SPECIFIC STRUCTURE PRESERVING HASHING FOR CROSS-MODAL RETRIEVAL
作者:Liu, Xingbo; Nie, Xiushan; Sun, Haoliang; Cui, Chaoran; Yin, Yilong
通讯作者:Nie, XS;Yin, YL;Nie, XS
作者机构:[Liu, Xingbo; Nie, Xiushan; Sun, Haoliang; Yin, Yilong] Shandong Univ, Sch Comp Sci & Technol, Jinan, Shandong, Peoples R China.; [Nie, Xiushan; Cui 更多
会议名称:IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
会议日期:APR 15-20, 2018
来源:2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
出版年:2018
页码:1678-1682
关键词:Cross-modal retrieval; Hashing; Modality-specific structure preserving;; Label enhancement
摘要:Hashing-based methods have made great advancements in cross-modal retrieval in both computational efficiency and storage. Learning a common space from different modalities is the common strategy of hashing-based methods, however, relational and structural information between samples in each modality, namely, a modality-specific structure, is always discarded during learning. In addition, cross-modality samples sometimes suffer from inter-class ambiguity and intra-class variability because of the uncertainty of manual labeling. To address these issues, we propose a novel method named Modality-specific structure Preserving Hashing (MsPH), which learns hashes by preserving the local structure and relations between samples in each modality. Moreover, label enhancement is utilized in MsPH to address label ambiguity and variability. Extensive experiments conducted on three benchmark datasets demonstrate the superiority of MsPH under various cross-modal scenarios.
收录类别:CPCI-S
资源类型:会议论文
TOP