Perceptual Loss based Approach for Analogue Film Restoration

Ivanova, D., Siebert, J. P. and Williamson, J. (2022) Perceptual Loss based Approach for Analogue Film Restoration. In: 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 6-8 Feb 2022, pp. 126-135. ISBN 9789897585555 (doi: 10.5220/0010829300003124)

Abstract

Analogue film restoration, both for still photographs and motion picture emulsions, is a slow and laborious manual process. Artifacts such as dust and scratches are random in shape, size, and location; additionally, the overall degree of damage varies between different frames. We address this less popular case of image restoration by training a U-Net model with a modified perceptual loss function. Along with the novel perceptual loss function used for training, we propose a more rigorous quantitative model evaluation approach which measures the overall degree of improvement in perceptual quality over our test set.

• Item Type: Conference Proceedings
• Additional Information: This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) [grant number EP/RS13222/1].
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: Ivanova, Ms Daniela and Siebert, Dr Paul and Williamson, Dr John
• Authors: Ivanova, D., Siebert, J. P., and Williamson, J.
• College/School: College of Science and Engineering > School of Computing Science
• ISBN: 9789897585555
• Published Online: 01 January 2022
• Copyright Holders: Copyright © 2022 by SCITEPRESS – Science and Technology Publications
• First Published: First published in Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - (Volume 4): 126-135
• Publisher Policy: Reproduced under a Creative Commons License

Related URLs

• Organisation
• Publisher

Project Code	Award No	Project Name	Principal Investigator	Funder's Name	Funder Ref	Lead Dept
305200		DTP 2018-19 University of Glasgow	Mary Beth Kneafsey	Engineering and Physical Sciences Research Council (EPSRC)	EP/R513222/1	MVLS - Graduate School