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

Super-resolution tasks oriented to images captured in ultra-dark environmentsis a practical yet challenging problem that has received little attention. Dueto uneven illumination and low signal-to-noise ratio in dark environments, amultitude of problems such as lack of detail and color distortion may bemagnified in the super-resolution process compared to normal-lightingenvironments. Consequently, conventional low-light enhancement orsuper-resolution methods, whether applied individually or in a cascaded mannerfor such problem, often encounter limitations in recovering luminance, colorfidelity, and intricate details. To conquer these issues, this paper proposes aspecialized dual-modulated learning framework that, for the first time,attempts to deeply dissect the nature of the low-light super-resolution task.Leveraging natural image color characteristics, we introduce a self-regularizedluminance constraint as a prior for addressing uneven lighting. Expanding onthis, we develop Illuminance-Semantic Dual Modulation (ISDM) components toenhance feature-level preservation of illumination and color details. Besides,instead of deploying naive up-sampling strategies, we design theResolution-Sensitive Merging Up-sampler (RSMU) module that brings togetherdifferent sampling modalities as substrates, effectively mitigating thepresence of artifacts and halos. Comprehensive experiments showcases theapplicability and generalizability of our approach to diverse and challengingultra-low-light conditions, outperforming state-of-the-art methods with anotable improvement (i.e., $\uparrow$5\% in PSNR, and $\uparrow$43\% in LPIPS).Especially noteworthy is the 19-fold increase in the RMSE score, underscoringour method's exceptional generalization across different darkness levels. Thecode will be available online upon publication of the paper.

Diving into Darkness: A Dual-Modulated Framework for High-Fidelity Super-Resolution in Ultra-Dark Environments