Denoising diffusion post-processing for low-light image enhancement

We are excited to announce that the repository will soon undergo an upgrade, featuring a new look and feel along with several enhanced features to improve your experience. Please be on the lookout for further updates and announcements regarding the launch date. We appreciate your support and look forward to unveiling the improved platform soon.

Show simple item record

dc.contributor.author Panagiotou, Savvas
dc.contributor.author Bosman, Anna
dc.date.accessioned 2024-08-26T07:59:21Z
dc.date.available 2024-08-26T07:59:21Z
dc.date.issued 2024-12
dc.description DATA AVAILABILITY : The public link to the code is specified in the manuscript. The datasets are listed in the manuscript and are open source/widely available. en_US
dc.description.abstract Low-light image enhancement (LLIE) techniques attempt to increase the visibility of images captured in low-light scenarios. However, as a result of enhancement, a variety of image degradations such as noise and color bias are revealed. Furthermore, each particular LLIE approach may introduce a different form of flaw within its enhanced results. To combat these image degradations, post-processing denoisers have widely been used, which often yield oversmoothed results lacking detail. We propose using a diffusion model as a post-processing approach, and we introduce Low-light Post-processing Diffusion Model (LPDM) in order to model the conditional distribution between under-exposed and normally-exposed images. We apply LPDM in a manner which avoids the computationally expensive generative reverse process of typical diffusion models, and post-process images in one pass through LPDM. Extensive experiments demonstrate that our approach outperforms competing post-processing denoisers by increasing the perceptual quality of enhanced low-light images on a variety of challenging low-light datasets. Source code is available at https://github.com/savvaki/LPDM. en_US
dc.description.department Computer Science en_US
dc.description.librarian hj2024 en_US
dc.description.sdg SDG-09: Industry, innovation and infrastructure en_US
dc.description.sponsorship The National Research Foundation (NRF) of South Africa. en_US
dc.description.uri https://www.elsevier.com/locate/pr en_US
dc.identifier.citation Panagiotou, S. & Bosman, A.S. 2024, 'Denoising diffusion post-processing for low-light image enhancement', Pattern Recognition, vol. 156, art. 110799, pp. 1-13, doi : 10.1016/j.patcog.2024.110799. en_US
dc.identifier.issn 0031-3203 (print)
dc.identifier.issn 1873-5142 (online)
dc.identifier.other 10.1016/j.patcog.2024.110799
dc.identifier.uri http://hdl.handle.net/2263/97851
dc.language.iso en en_US
dc.publisher Elsevier en_US
dc.rights © 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC license. en_US
dc.subject Low-light image enhancement (LLIE) en_US
dc.subject Diffusion model en_US
dc.subject Denoising en_US
dc.subject Post-processing en_US
dc.subject SDG-09: Industry, innovation and infrastructure en_US
dc.subject Low-light post-processing diffusion model (LPDM) en_US
dc.title Denoising diffusion post-processing for low-light image enhancement en_US
dc.type Article en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record