AI Paper Atlas

Research map / Image generation / Super-resolution

Emulating Self-attention with Convolution for Efficient Image Super-Resolution

Dongheon Lee, Seokju Yun, Youngmin Ro

ICCV 2025 · Image generation › Super-resolution · approach: CNN and transformer SR architectures · 42 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 ConvAttn layers to a self-attention-only network (ablation)ESC · Urban100 x2PSNR: 33.23 -> 33.46Improved
add self-attention (first layer per block) to a ConvAttn-only network (ablation)ESC · Urban100 x2PSNR: 32.91 -> 33.46Improved
use 13x13 shared large kernel instead of 9x9 (ablation)ESC · Urban100 x2PSNR: 33.42 -> 33.46Improved
use 13x13 shared large kernel instead of 17x17 (ablation)ESC · Urban100 x2PSNR: 33.40 -> 33.46Improved
add LK sharing and dynamic kernel (ablation)ESC · Urban100 x2PSNR: 33.37 -> 33.46Improved
add LK sharing (ablation)ESC · Urban100 x2PSNR: 33.43 -> 33.46Improved
add dynamic kernel DK (ablation)ESC · Urban100 x2PSNR: 33.36 -> 33.46Improved
enlarge self-attention window to 32 from 16 with more layers (ablation)ESC · Urban100 x2PSNR: 33.05 -> 33.46Improved
enlarge self-attention window to 32 from 24 with more layers (ablation)ESC · Urban100 x2PSNR: 33.28 -> 33.46Improved
use relative position bias with Flash Attention instead of RoPE (ablation)ESC · Urban100 x2PSNR: 32.76 -> 33.46Improved
add Flash Attention to 32x32 window self-attentionself-attention · 288x288x64 inputlatency: 44.7 -> 2.8Improved
add Flash Attention to 32x32 window self-attentionself-attention · 288x288x64 inputmemory (MB): 2717 -> 223Improved

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