AI Paper Atlas

Research map / Image generation / Super-resolution

Unsupervised Real-World Image Super Resolution via Domain-Distance Aware Training

Yunxuan Wei, Shuhang Gu, Yawei Li, Radu Timofte, Longcun Jin, Hengjie Song

CVPR 2021 · Image generation › Super-resolution · approach: Real-world super-resolution · 161 citations (OpenAlex)

Check 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 domain-gap aware target-domain adversarial training (ablation)DASR SR network · AIMLPIPS: 0.378 -> 0.359Improved
Add domain-gap aware target-domain adversarial training (ablation)Bicubic-source SR network · AIMLPIPS: 0.543 -> 0.390Improved
Add domain-distance weighted supervision (ablation)DASR SR network · AIMLPIPS: 0.359 -> 0.336Improved
Add domain-distance weighted supervision (ablation)Generated-pair SR network without target training · AIMLPIPS: 0.378 -> 0.348Improved
Combine target-domain training and weighted supervision (ablation)DASR SR network · AIMLPIPS: 0.378 -> 0.336Improved
Use HR images directly rather than bicubic LR as DSN input (ablation)Down-sampling network · AIM validationLPIPS: 0.110 -> 0.101Improved
Use wavelet-frequency adversarial loss instead of GBFS (ablation)Down-sampling network with HR input · AIM validationLPIPS: 0.101 -> 0.067Improved
Use wavelet-frequency adversarial loss instead of RGB adversarial loss (ablation)Down-sampling network with HR input · AIM validationLPIPS: 0.138 -> 0.067Improved
Use DASR rather than ZSSRReal-world image SR · AIMLPIPS: 0.630 -> 0.336Improved
Use DASR rather than FSSRReal-world image SR · RealSRPSNR: 25.992 -> 26.782Improved
Use DASR rather than FSSRReal-world image SR · CameraSRLPIPS: 0.180 -> 0.151Improved
Use DASR rather than supervised ESRGANReal-world image SR · RealSRLPIPS: 0.199 -> 0.228Hurt

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