AI Paper Atlas

Research map / Image generation / Image restoration

FAPE-IR: Frequency-Aware Planning and Execution Framework for All-in-One Image Restoration

Jingren Liu, Shuning Xu, Qirui Yang, Yun Wang, Xiangyu Chen, Zhong Ji

CVPR 2026 · Image generation › Image restoration · approach: Efficient restoration architectures · 3 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 Qwen2.5-VL planner without routing control (ablation)FAPE-IR executor · URHIPSNR: 25.03 -> 27.95Improved
add Freq-U planner-side text router (ablation)FAPE-IR · URHIPSNR: 27.95 -> 28.92Improved
add Freq-G FIR spectral router (ablation)FAPE-IR · URHIPSNR: 28.92 -> 29.71Improved
use mildly asymmetric ranks 8/16 with Freq-G (ablation)FAPE-IR · URHIPSNR: 29.71 -> 29.60No clear effect
use extremely skewed ranks 4/16 (ablation)FAPE-IR · URHIPSNR: 29.71 -> 25.93Hurt
FAPE-IR vs best baseline on derainingFAPE-IR · DerainingPSNR: 21.94 -> 28.30Improved
FAPE-IR vs best baseline on denoisingFAPE-IR · DenoisingPSNR: 31.63 -> 30.34Hurt
FAPE-IR vs DFPIR on low-lightFAPE-IR · Low-light enhancementPSNR: 25.92 -> 25.34Hurt
FAPE-IR vs PASD on SRFAPE-IR · Super-resolutionPSNR: 26.87 -> 28.53Improved
FAPE-IR vs PURE inference timeFAPE-IR · Complexity comparisontime (s): 201.67 -> 1.57Improved
FAPE-IR vs lightweight AIO-IR methods on inference timeFAPE-IR · Complexity comparisontime (s): 0.08 -> 1.57Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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