AI Paper Atlas

Research map / Image generation / Image restoration

Decouple Learning for Parameterized Image Operators

Qingnan Fan, Dongdong Chen, Lu Yuan, Gang Hua, Nenghai Yu, Baoquan Chen

ECCV 2018 · Image generation › Image restoration · approach: Priors for restoration networks · 65 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
train one network jointly on continuous random parameter values instead of one per valueL0 filter network · L0 smoothingPSNR diff: 0.86Hurt
train jointly on random parameters instead of per-value networksWLS filter network · WLS filterPSNR diff: 0.61Hurt
train jointly on random parameters instead of per-value networksRTV filter network · RTV filterPSNR diff: 0.31Hurt
train jointly on random parameters instead of per-value networksSuper resolution network · Super resolutionPSNR: 29.29 -> 29.23No clear effect
train jointly on random parameters instead of per-value networksDenoising network · DenoisingPSNR: 28.77 -> 28.73No clear effect
train jointly on random parameters instead of per-value networksDeblock network · DeblockPSNR: 30.37 -> 30.30No clear effect
use two-dimensional parameters (lambda, sigma) in one network, varying sigmaRTV network · RTV (lambda=0.01)PSNR: 40.86 -> 40.93No clear effect
use two-dimensional parameters (lambda, sigma) in one network, varying lambdaRTV network · RTV (sigma=3)PSNR: 40.86 -> 40.14Hurt
joint training over 6 filters or 4 restoration tasks instead of one operator (ablation)Decouple learning · 10 image operatorsPSNR (avg): 34.86 -> 33.49Hurt
joint training over all 10 operators instead of one operator (ablation)Decouple learning · 10 image operatorsPSNR (avg): 34.86 -> 32.89Hurt
joint training over all 10 operators instead of one operator (ablation)Decouple learning · 10 image operatorsSSIM (avg): 0.925 -> 0.914Hurt
disable parameter specification for restoration tasks in joint trainingDecouple learning · restoration tasksPSNRNo clear effect

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