Image generation from text with entity information fusion

作者：

Highlights：

• Two end-to-end frameworks to incorporate entity information for image generation have been proposed.

• A novel evaluation metric, Entity Matching Score (EMS), has been proposed.

• Experimental results show that the proposed approach outperforms the existing models.

摘要

•Two end-to-end frameworks to incorporate entity information for image generation have been proposed.•A novel evaluation metric, Entity Matching Score (EMS), has been proposed.•Experimental results show that the proposed approach outperforms the existing models.

论文关键词：00-01,99-00,Image generation from text,Entity information fusion,End-to-end frameworks,Entity Matching Score

论文评审过程：Received 11 January 2021, Revised 21 May 2021, Accepted 2 June 2021, Available online 6 June 2021, Version of Record 18 June 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107200