AI Paper Atlas

Research map / Image generation / Super-resolution

Efficient Real-World Image Super-Resolution Via Adaptive Directional Gradient Convolution

Long Peng, Zhanfeng Feng, Renjing Pei, Wenbo Li, Jiaming Guo, Xueyang Fu, Yang Wang, Yang Hua Cao, Zheng-Jun Zha

IJCV 2026 · Image generation › Super-resolution · approach: CNN and transformer SR architectures · 40 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
replace VConv with DGConvVDSR · Real-SR avgPSNR: 23.65 -> 24.14Improved
replace VConv with DGConvEDSR · Real-SR avgPSNR: 24.99 -> 25.18Improved
replace VConv with DGConvIMDN · Real-SR avgPSNR: 24.11 -> 24.52Improved
replace VConv with DGConvSRResNet · Real-SR avgLPIPS: 0.352 -> 0.335Improved
use five parallel VConvs instead of one (ablation)SRResNet · Real-SR avgPSNR: 24.996 -> 24.77Hurt
use ECB block instead of VConv (ablation)SRResNet · Real-SR avgPSNR: 24.996 -> 24.85Hurt
add IDG convolution (ablation)DGConv · Real-SR avgPSNR: 25.21 -> 25.41Improved
add CSG convolution (ablation)DGConv · Real-SR avgLPIPS: 0.323 -> 0.309Improved
add VConv branch (ablation)DGConv · Real-SR avgPSNR: 25.38 -> 25.41Hurt
increase feature channels 32 -> 64 (ablation)DGPNet · Real-SR avgPSNR: 25.37 -> 25.41Improved
use local mean instead of central value (ablation)DGConv on EDSR · Real-SR avgPSNR: 24.947 -> 25.186Improved
replace learnable convolution with fixed handcrafted kernels (ablation)SRResNet · Real-SRPSNR: 0.87 dB dropHurt

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