AI Paper Atlas

Research map / Image generation / Super-resolution

LIIFusion: Coarse-to-fine Framework for Generative MEF via Implicit Neural Representation

Sangmin Han, Jinho Kim, Jinwoo Kim, Dongyoung Kim, Seon Joo Kim

ECCV 2026 · Image generation › Super-resolution · approach: Diffusion-based super-resolution

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
replace patch-wise diffusion at HR with low-res diffusion plus LIIF fine stageUltraFusion · UltraFusion Benchmarktotal inference time: 203 min -> 59 minImproved
coarse-to-fine design reduces computeUltraFusion · 1988x1326 imageTFLOPs: 944.4 -> 282.1Improved
LIIFusion vs UltraFusion on perceptual qualityUltraFusion · UltraFusion BenchmarkMUSIQ/HyperIQA: MUSIQ 68.40 -> 70.19; HyperIQA 0.6214 -> 0.6467Improved
LIIFusion vs UltraFusion on DeQA-ScoreUltraFusion · UltraFusion BenchmarkDeQA-Score: 4.0123 -> 3.9807Hurt
LIIFusion vs UltraFusion on structural MEF-SSIMUltraFusion · MEFBMEF-SSIM: 0.9266 -> 0.9201Hurt
add adaptive exposure correction (alpha 0.2) (ablation)coarse stage · MEFBMEF-SSIM: 0.9266 -> 0.9313Improved
add adaptive exposure correction (alpha 0.2) (ablation)coarse stage · MEFBMUSIQ: 68.14 -> 67.91Hurt
raise AEC alpha to 1.0 (ablation)coarse stage · MEFBMEF-SSIM: 0.9313 -> 0.8633Hurt
train LIIF for multi-exposure fusion instead of using SR-pretrained LIIF (ablation)fine stage · MEFBMEF-SSIM, HyperIQAImproved
feed both OE and UE inputs to fine stage (ablation)fine stage · UltraFusion BenchmarkNRIQA metricsImproved
train on UltraFusion pseudo labels instead of SICE labels (ablation)LIIFusion · UltraFusion BenchmarkNRIQA metricsNo clear effect
add ICCV23 pseudo-labeled data to training (ablation)LIIFusion · UltraFusion BenchmarkMUSIQ/PAQ2PIQ/HyperIQA: MUSIQ 70.19 -> 70.22; PAQ2PIQ 75.59 -> 75.46; HyperIQA 0.6467 -> 0.6506Mixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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