AI Paper Atlas

Research map / Image generation / Super-resolution

Image Super-Resolution with Guarantees via Conformalized Generative Models

Eduardo Adame, Daniel Csillag, Guilherme Tegoni Goedert

NeurIPS 2025 · Image generation › Super-resolution · approach: Perceptual and real-world SR · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes conformal "confidence masks" for generative image super-resolution that guarantee expected fidelity error (and PSNR) control on any black-box model using unlabelled calibration images.

Problem

Quantifying where generated super-resolution images can be trusted, with interpretable, statistically guaranteed uncertainty for black-box generative models.

Approach

Calibrate a threshold on a smoothed per-pixel model-indecision map (variance computed via a low-pass kernel on second moments) using conformal risk control, so that the expected supremum of a local fidelity error over the trusted binary mask is at most alpha; also proves PSNR lower bounds and bounds under calibration data leakage.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Apply conformal masking at α=0.100SinSR super-resolution · Liu4K test, semantic D_pPSNR: 26.83 ± 1.06 -> 32.65 ± 1.48Improved
Apply conformal masking at α=0.100SinSR super-resolution · Liu4K test, non-semantic D_pPSNR: 26.82 ± 1.08 -> 28.64 ± 0.93Improved
Apply conformal masking at α=0.100SinSR super-resolution · Liu4K test, semantic D_pfidelity error: 0.14 ± 0.01 -> 0.03 ± 0.01Improved
Apply conformal masking at α=0.100SinSR super-resolution · Liu4K test, non-semantic D_pfidelity error: 0.14 ± 0.01 -> 0.09 ± 0.01Improved
Use a neighborhood-averaged, low-pass-filtered D_p instead of pointwise D_p (ablation)Conformal mask · Image super-resolutionconformal mask sizeImproved
Increase the Gaussian blur radius for D_p (ablation)Conformal mask · Image super-resolutiontrusted-region coverageImproved
Increase the Gaussian blur radius for σ (ablation)Conformal mask · Image super-resolutiontrusted-region coverageImproved
Use pixel-wise variance σ^var without patch-based low-pass filtering (ablation)Conformal mask · Image super-resolutionconformal mask qualityHurt
Calibrate on DIV2K distribution-shifted data instead of in-distribution Liu4KConformal mask · DIV2K validationPSNR: 13.34 ± 6.91; 13.14 ± 6.55; baseline 11.73 ± 0.39; 11.74 ± 0.39Mixed

Surprising findings

Limitations

Experimental setup

Datasets: Liu4K DIV2K

Benchmarks: Liu4K validation (400 images) DIV2K validation (distribution shift)

Metrics: fidelity error PSNR conformal mask size

Baselines: interval-valued conformal super-resolution method (unnamed in text) continuous confidence-mask method (unnamed in text) no uncertainty quantification

Builds on: conformal risk control conformal prediction

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