AI Paper Atlas

Research map / Image generation / Image restoration

Low-Light Image Enhancement with Normalizing Flow

Yufei Wang, Renjie Wan, Wenhan Yang, Haoliang Li, Lap-Pui Chau, Alex Kot

AAAI 2022 · Image generation › Image restoration · approach: Low-light image enhancement · 500 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use conditional normalizing flow with NLL training instead of L1 training (ablation)LLFlow · LOLPSNR: 22.68 -> 25.19Improved
Use conditional normalizing flow with NLL training instead of L1 training (ablation)LLFlow · LOLSSIM: 0.8391 -> 0.9252Improved
Use conditional normalizing flow with NLL training instead of L1 training (ablation)LLFlow · LOLLPIPS: 0.2038 -> 0.1131Improved
Add the conditional invertible flow to the encoder-only model (ablation)LLFlow · LOLPSNR: 21.90 -> 25.19Improved
Add the illumination-invariant color map prior (ablation)LLFlow · LOLPSNR: 24.46 -> 25.19Improved
Add the illumination-invariant color map prior (ablation)LLFlow · LOLSSIM: 0.9235 -> 0.9252Improved
Add the illumination-invariant color map prior (ablation)LLFlow · LOLLPIPS: 0.1146 -> 0.1131Improved
Set color-map selection probability p to 0.5 rather than 0.2 (ablation)LLFlow · LOLPSNR: 25.19 -> 24.85Hurt
Set color-map selection probability p to 0.5 rather than 0.2 (ablation)LLFlow · LOLSSIM: 0.9252 -> 0.9232Hurt
Set color-map selection probability p to 0.5 rather than 0.2 (ablation)LLFlow · LOLLPIPS: 0.1131 -> 0.1192Hurt
Use latent z values with different offsets (ablation)LLFlow · LOLBrightnessImproved
Use the learned NLL rather than pixel-wise L1 as the image-quality likelihood (ablation)LLFlow · LOL test image perturbationsNLL: -5.95; -6.15; 4.84Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy