English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, we rethink the low-light image enhancement task and propose aphysically explainable and generative diffusion model for low-light imageenhancement, termed as Diff-Retinex. We aim to integrate the advantages of thephysical model and the generative network. Furthermore, we hope to supplementand even deduce the information missing in the low-light image through thegenerative network. Therefore, Diff-Retinex formulates the low-light imageenhancement problem into Retinex decomposition and conditional imagegeneration. In the Retinex decomposition, we integrate the superiority ofattention in Transformer and meticulously design a Retinex Transformerdecomposition network (TDN) to decompose the image into illumination andreflectance maps. Then, we design multi-path generative diffusion networks toreconstruct the normal-light Retinex probability distribution and solve thevarious degradations in these components respectively, including darkillumination, noise, color deviation, loss of scene contents, etc. Owing togenerative diffusion model, Diff-Retinex puts the restoration of low-lightsubtle detail into practice. Extensive experiments conducted on real-worldlow-light datasets qualitatively and quantitatively demonstrate theeffectiveness, superiority, and generalization of the proposed method.

Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model