AI Paper Atlas

Research map / Image generation / Super-resolution

Learning Where to Look and How to Judge: Resolution-agnostic Image Quality Assessment with Quality-aware Saliency

Hakan Emre Gedik, Shashank Gupta, Alan Bovik

CVPR 2026 · Image generation › Super-resolution · approach: Perceptual and real-world SR · 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 PIE importance weighting instead of mean pooling of patches (ablation)ReLIQS · KonIQ-10KSRCC: 0.941 -> 0.949Improved
add PIE importance weighting instead of mean pooling of patches (ablation)ReLIQS · UHDSRCC: 0.861 -> 0.865Improved
add 512 scale to 224 scale (ablation)ReLIQS · UHDSRCC: 0.680 -> 0.750Improved
add original-resolution scale to 224 and 512 (ablation)ReLIQS · UHDSRCC: 0.750 -> 0.865Improved
add original-resolution scale to 224 and 512 (ablation)ReLIQS · KonIQ-10KSRCC: 0.938 -> 0.949Improved
add multiple scales on lower-resolution datasets (ablation)ReLIQS · SPAQ, CLIVE, KADID, CSIQ, LIVEPLCC/SRCC: 0.882 / 0.891 -> 0.891 / 0.894No clear effect
increase latent quality axes from 1 to 4 (ablation)ReLIQS · UHDSRCC: 0.858 -> 0.859 -> 0.865Improved
increase latent quality axes from 4 to 8 (ablation)ReLIQS · UHDSRCC: 0.865 -> 0.863No clear effect
add margin-ranking loss to PLCC loss (ablation)ReLIQS · KonIQ-10KSRCC: 0.941 -> 0.949Improved
add PLCC loss to margin-ranking loss (ablation)ReLIQS · KonIQ-10KPLCC: 0.924 -> 0.958Improved
use uncertainty-based loss weighting instead of equal weights (ablation)ReLIQS · KonIQ-10K, SPAQ, KADID multi-datasetSRCC: 0.944 -> 0.946Improved
select only top-48 of 748 original-resolution patches by PIE weightReLIQS · UHDSRCC: 748 -> 48 patchesNo clear effect

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