Cross-modal learning with prior visual relation knowledge

作者：

Highlights：

• Human prior knowledge facilitates visual relational reasoning.

• Anisotropic Graph Convolution can generate relation-aware image representation.

• Relational Reasoning module is plug-and-play.

摘要

•Human prior knowledge facilitates visual relational reasoning.•Anisotropic Graph Convolution can generate relation-aware image representation.•Relational Reasoning module is plug-and-play.

论文关键词：Visual relation reasoning,Relation embedding,Anisotropic graph convolutional networks,Visual question answering,Cross-modal information retrieval

论文评审过程：Received 29 January 2020, Revised 11 May 2020, Accepted 14 June 2020, Available online 16 June 2020, Version of Record 23 June 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2020.106150