AI Paper Atlas

Research map / Image generation / Super-resolution

Feature Modulation Transformer: Cross-Refinement of Global Representation via High-Frequency Prior for Image Super-Resolution

Ao Li, Le Zhang, Yun Liu, Ce Zhu

ICCV 2023 · Image generation › Super-resolution · approach: Efficient lightweight super-resolution · 155 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 SRWAB (transformer global branch) to HFERB-only model (ablation)CRAFT · Manga109 x4PSNR: 30.79 -> 31.18Improved
add HFERB high-frequency prior to SRWAB-only model (ablation)CRAFT · Manga109 x4PSNR: 31.12 -> 31.18Improved
use HFB inter-attention fusion instead of concatenation (ablation)CRAFT · Manga109 x4PSNR: 30.92 -> 31.18Improved
use HFERB output as query and SRWAB as key/value rather than swapped (ablation)CRAFT · Manga109 x4PSNR: 30.67 -> 31.18Improved
use parallel high-frequency prior fusion instead of cascade (ablation)CRAFT · Manga109 x4PSNR: 30.88 -> 31.18Improved
CRAFT architecture instead of SwinIRSISR x4 · 128x128 inputParams / FLOPs / time: 897K -> 753K params; 72.0 -> 42.8 msImproved

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