Discriminative correlation hashing for supervised cross-modal retrieval
作者：
Highlights：
•
• We propose a novel cross-modal hashing framework DCH, which integrates discriminative properties into the hashing learning procedure.
• The linear discriminant analysis is introduced on textual modality to preserve the discriminative distribution of text data points.
• We use the unified binary codes to represent two modalities in one instance which effectively reduce semantic gap and transfer the discriminative characteristic to visual modality.
• Experimental results demonstrate the effectiveness of the proposed approach.
摘要
•We propose a novel cross-modal hashing framework DCH, which integrates discriminative properties into the hashing learning procedure.•The linear discriminant analysis is introduced on textual modality to preserve the discriminative distribution of text data points.•We use the unified binary codes to represent two modalities in one instance which effectively reduce semantic gap and transfer the discriminative characteristic to visual modality.•Experimental results demonstrate the effectiveness of the proposed approach.
论文关键词：Cross-modal retrieval,Hashing,Subspace learning,Discriminant analysis
论文评审过程：Received 5 June 2017, Revised 10 April 2018, Accepted 11 April 2018, Available online 17 April 2018, Version of Record 7 May 2018.
论文官网地址：https://doi.org/10.1016/j.image.2018.04.009
