AI Paper Atlas

Research map / Image generation / Super-resolution

Learning Many-to-Many Mapping for Unpaired Real-World Image Super-Resolution and Downscaling

Wanjie Sun, Zhenzhong Chen

TPAMI 2024 · Image generation › Super-resolution · approach: Real-world super-resolution · 17 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 backward loss (perceptual and GAN) at tau=0SDFlow · RealSR 4x SRLPIPS: 0.2934 -> 0.2500Improved
add backward loss (perceptual and GAN) at tau=0.8SDFlow · RealSR 4x SRNIQE: 5.7145 -> 4.8106Improved
add backward loss at tau=0.8SDFlow · DRealSR 4x SRNIQE: 7.6900 -> 5.5167Improved
raise sampling temperature from 0 to 0.8 (forward loss)SDFlow · RealSR 4x SRLPIPS: 0.2934 -> 0.2321Improved
raise sampling temperature from 0 to 0.8 (forward loss)SDFlow · RealSR 4x SRPSNR: 28.5813 -> 27.8403Hurt
raise sampling temperature from 0 to 0.8 (forward loss)SDFlow · DRealSR 4x SRLPIPS: 0.3411 -> 0.2852Improved
add backward loss at tau=0.8SDFlow · DRealSR 4x SRPSNR: 30.5901 -> 29.4299Hurt
use SDFlow instead of ADL for downscalingdownscaling model · RealSR 4x downscalingPSNR: 34.5064 -> 34.8498Improved
use SDFlow instead of DeFlow for synthesizing LR training pairsESRGAN · RealSR 4x SR (ESRGAN trained on synthesized pairs)LPIPS: 0.3356 -> 0.2262Improved
use SDFlow instead of DeFlow for synthesizing LR training pairsESRGAN · RealSR 4x SR (ESRGAN trained on synthesized pairs)NIQE: 7.2015 -> 5.2521Improved
SDFlow vs DeFlow downscaling diversitydownscaling model · RealSR 4x downscalingDiversity: 10.1595 -> 2.2314Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 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