Which is the better inpainted image? Learning without subjective annotation

Mariko Isogawa, Dan Mikami, Kosuke Takahashi, Hideaki Kimata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)


This paper proposes a learning-based quality evaluation framework for inpainted results that does not require any subjectively annotated training data. Image inpainting, which removes and restores unwanted regions in images, is widely acknowledged as a task whose results are quite difficult to evaluate objectively. Thus, existing learning-based image quality assessment (IQA) methods for inpainting require subjectively annotated data for training. However, subjective annotation requires huge cost and subjects’ judgment occasionally differs from person to person in accordance with the judgment criteria. To overcome these difficulties, the proposed framework uses simulated failure results of inpainted images whose subjective qualities are controlled as the training data. This approach enables preference order between pairwise inpainted images to be successfully estimated even if the task is quite subjective. To demonstrate the effectiveness of our approach, we test our algorithm with various datasets and show it outperforms state-of-the-art IQA methods for inpainting.

Original languageEnglish
Title of host publicationBritish Machine Vision Conference 2017, BMVC 2017
PublisherBMVA Press
ISBN (Electronic)190172560X, 9781901725605
Publication statusPublished - 2017
Externally publishedYes
Event28th British Machine Vision Conference, BMVC 2017 - London, United Kingdom
Duration: 2017 Sept 42017 Sept 7

Publication series

NameBritish Machine Vision Conference 2017, BMVC 2017


Conference28th British Machine Vision Conference, BMVC 2017
Country/TerritoryUnited Kingdom

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition


Dive into the research topics of 'Which is the better inpainted image? Learning without subjective annotation'. Together they form a unique fingerprint.

Cite this