A composition-oriented aesthetic view recommendation network supervised by the simplified golden ratio theory

作者：

Highlights：

• A view recommendation model constrained by aesthetic evaluation and composition.

• A Simplified Golden Ratio Theory-based candidate box extraction.

• A dataset is constructed using the proposed candidate box selection strategy.

• An aesthetic evaluation network and a multitask loss function are proposed.

摘要

•A view recommendation model constrained by aesthetic evaluation and composition.•A Simplified Golden Ratio Theory-based candidate box extraction.•A dataset is constructed using the proposed candidate box selection strategy.•An aesthetic evaluation network and a multitask loss function are proposed.

论文关键词：Aesthetic view recommendation,Image aesthetic quality assessment,Composition classification,Image cropping,Distillation model

论文评审过程：Received 4 August 2021, Revised 25 September 2021, Accepted 1 January 2022, Available online 29 January 2022, Version of Record 9 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116500