.png){kind=icon}
.png){kind=icon}
Return
Learning interpretable binary codes via semantic alignment for customized image retrieval
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2025.112380.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Propose to endow hash bits with low/middle level semantics. • Propose an implicit semantic alignment scheme to transfer the semantic knowledge of filters to target hash bits. • Introduce a shared classification rule matrix to disentangle the functionalities of convolutional filters and hash bits. • Demonstrate promising performance of the proposed method for both traditional and customized image retrieval tasks.
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W