Privileged multi-task learning for attribute-aware aesthetic assessment
作者：
Highlights：
• We propose the first unified approach to model the multiple complex dependencies in a photo for aesthetic assessment.
• We employ privileged information during training and incorporate auxiliary aesthetics photo features to assist aesthetics prediction within a deep learning architecture.
• We propose to employ adversarial learning to serves as an additional view to refine the final aesthetics assessment performance.
摘要
•We propose the first unified approach to model the multiple complex dependencies in a photo for aesthetic assessment.•We employ privileged information during training and incorporate auxiliary aesthetics photo features to assist aesthetics prediction within a deep learning architecture.•We propose to employ adversarial learning to serves as an additional view to refine the final aesthetics assessment performance.
论文关键词：Aesthetic assessment,Privileged information,Multi-task learning
论文评审过程：Received 29 July 2021, Revised 3 June 2022, Accepted 21 July 2022, Available online 22 July 2022, Version of Record 29 July 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108921
