Deep Discrete Cross-Modal Hashing for Cross-Media Retrieval

作者：

Highlights：

• A framework of cross-modal deep neural networks is proposed for cross- media retrieval.

• Intra-modality and inter-modality similarity preserving are considered in binary codes learning.

• Binary constraint is addressed by discrete cyclic coordinate descent (DCC) algorithm without relaxation.

• Experimental results on four datasets demonstrate the effectiveness of the proposed DDCMH, which is significantly superior to state-of-the-art cross-modal hashing approaches.

摘要

•A framework of cross-modal deep neural networks is proposed for cross- media retrieval.•Intra-modality and inter-modality similarity preserving are considered in binary codes learning.•Binary constraint is addressed by discrete cyclic coordinate descent (DCC) algorithm without relaxation.•Experimental results on four datasets demonstrate the effectiveness of the proposed DDCMH, which is significantly superior to state-of-the-art cross-modal hashing approaches.

论文关键词：Cross-modal retrieval,deep learning,discrete hashing,alternative optimization

论文评审过程：Received 30 November 2017, Revised 2 May 2018, Accepted 20 May 2018, Available online 21 May 2018, Version of Record 26 May 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.05.018