AI Paper Atlas

Research map / Image generation / Super-resolution

Accelerating Image Super-Resolution Networks with Pixel-Level Classification

Jinho Jeong, Jinwoo Kim, Younghyun Jo, Seon Joo Kim

ECCV 2024 · Image generation › Super-resolution · approach: CNN and transformer SR architectures · 10 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 patch-level allocation with pixel-level classificationCARN · Test2K x4GFLOPs: 99.8 (ARM) -> 64.0 at similar PSNRImproved
increase patch size for patch-level ClassSR versus PCSR (ablation)CARN-ClassSR · Test2K x4GFLOPs: 66.7 -> 75.8Hurt
increase patch size for PCSR (ablation)CARN-PCSR · Test2K x4GFLOPs: 61.1 -> 54.5Improved
use 3 classes instead of 2 (ablation)CARN-PCSR · Test2K x4GFLOPs: 64.0 -> 62.4No clear effect
add pixel-wise refinement (#h=1) to reduce artifacts (ablation)CARN-PCSR · Test2K x4PSNR: 26.022 -> 25.995 (#h=9 vs #h=0)No clear effect
use LIIF upsampler instead of pixel-shuffle (ablation)SRResNet · Urban100 x4PSNR: 24.65 -> 24.63No clear effect
apply PCSR to BSRN lightweight modelBSRN · Urban100 x4PSNR: 24.43 -> 24.29, GFLOPs 29.1 -> 23.6Mixed
compare to per-image method MGAFSRCNN-PCSR · Test4K x4GFLOPs: 101.7 -> 37.8Improved
apply PCSR on small images (Urban100) with ClassSR/ARMFSRCNN-ClassSR · Urban100 x4GFLOPs: 19.9 -> 20.8 (105%)Hurt

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