AI Paper Atlas

Research map / Image generation / Super-resolution

Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks

Cheeun Hong, Kyoung Mu Lee

ECCV 2024 · Image generation › Super-resolution · approach: CNN and transformer SR architectures · 4 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 weight clipping correction (ablation)ODM (2-bit EDSR) · Set5PSNR: 29.94 -> 31.12Improved
naively add mismatch regularization to reconstruction loss (ablation)ODM (2-bit EDSR) · Set5PSNR: 31.12 -> 30.99Hurt
weigh mismatch regularization gradient by cosine similarity with reconstruction gradient (cooperative) (ablation)ODM (2-bit EDSR) · Set5PSNR: 30.99 -> 31.50Improved
add cooperative mismatch regularization without weight clipping correction (ablation)ODM (2-bit EDSR) · Set5PSNR: 29.94 -> 30.34Improved
ODM in place of dynamic quantization modules (DAQ, DDTB)EDSR 2-bit · Set5PSNR: 31.01 -> 31.50Improved
ODM on RDN 2-bitRDN · Set5PSNR: 30.71 -> 31.37Improved
ODM on SwinIR 2-bit versus QuantSR (300K iterations)SwinIR · Set5PSNR: 31.53 -> 31.67Improved
ODM versus DAQ computation overheadEDSR 2-bit · complexityBitOPs: 215.6T -> 215.1TImproved

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