Learning Image Aesthetics by Learning Inpainting


June, Hao Ching and See, John Su Yang and Wong, Lai Kuan (2020) Learning Image Aesthetics by Learning Inpainting. In: 2020 IEEE International Conference on Image Processing (ICIP), 25-28 Oct. 2020, Virtual, Abu Dhabi, UAE.

[img] Text

Download (147kB)


Due to the high capability of learning robust features, convolutional neural networks (CNN) are becoming a mainstay solution for many computer vision problems, including aesthetic quality assessment (AQA). However, there remains the issue that learning with CNN requires time-consuming and expensive data annotations especially for a task like AQA. In this paper, we present a novel approach to AQA that incorporates self-supervised learning (SSL) by learning how to inpaint images according to photographic rules such as rules-of-thirds and visual saliency. We conduct extensive quantitative experiments on a variety of pretext tasks and also different ways of masking patches for inpainting, reporting fairer distribution-based metrics. We also show the suitability and practicality of the inpainting task which yielded comparably good benchmark results with much lighter model complexity

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Neural networks (Computer science), Aesthetic quality assessment, CNN, self-supervised learning, image inpainting, photographic rules
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Rosnani Abd Wahab
Date Deposited: 10 Sep 2021 15:13
Last Modified: 10 Sep 2021 15:13
URII: http://shdl.mmu.edu.my/id/eprint/8518


Downloads per month over past year

View ItemEdit (login required)