AI Paper Atlas

Research map / Image generation / Image restoration

Learning Underwater Image Enhancement Iteratively Without Reference Images

Yi Tang, Hiroshi Kawasaki, Takafumi Iwaguchi, Yuhang Zhang, Hiroshi Masui

AAAI 2026 · Image generation › Image restoration · approach: Inverse imaging with learned priors · 1 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
add style branch (ablation)unsupervised diffusion UIE · LSUIUIQM: 3.121 -> 3.508Improved
add quantification-mechanism feature weights (ablation)unsupervised diffusion UIE · LSUIUIQM: 3.256 -> 3.508Improved
add color filter on conditional image (ablation)unsupervised diffusion UIE · LSUIURanker: 1.904 -> 2.249Improved
add post-processing module (ablation)unsupervised diffusion UIE · LSUIUIQM: 3.318 -> 3.508Improved
add text prompts (ablation)unsupervised diffusion UIE · LSUIUIQM: 3.322 -> 3.508Improved
iterate training from round 1 to round 5 (ablation)unsupervised diffusion UIE · LSUIUIQM: 3.115 -> 3.508Improved
add a sixth training round (ablation)unsupervised diffusion UIE · LSUIUIQM: 3.508 -> 3.479Hurt
compare unsupervised method to supervised WF-DiffOurs · LSUIPSNR: 27.26 vs 18.42Hurt
compare unsupervised method to supervised models on no-reference metricsOurs · UIEB C60UIQM: 3.251 vs 3.044 (UIEDAL)Improved

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