Cross-modal image–text search via Efficient Discrete Class Alignment Hashing

作者：

Highlights：

• The EDCAH improves the discriminative capability of the hash codes.

• The EDCAH-t reduces the training complexity of the proposed model.

• A discrete optimization strategy efficiently solves the EDCAH model.

• Analysis shows the superiority of EDCAH and EDCAH-t in accuracy and efficiency.

摘要

•The EDCAH improves the discriminative capability of the hash codes.•The EDCAH-t reduces the training complexity of the proposed model.•A discrete optimization strategy efficiently solves the EDCAH model.•Analysis shows the superiority of EDCAH and EDCAH-t in accuracy and efficiency.

论文关键词：Class alignment,Cross-modal image–text search,Hash code,Supervised hashing

论文评审过程：Received 28 June 2021, Revised 4 January 2022, Accepted 22 January 2022, Available online 7 February 2022, Version of Record 7 February 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.102886