AI Paper Atlas

Research map / Optimization / Optimizers

Self-Supervised Deep Image Restoration via Adaptive Stochastic Gradient Langevin Dynamics

Weixi Wang, Ji Li, Hui Ji

CVPR 2022 · Optimization › Optimizers · approach: Stochastic gradient MCMC · 19 citations (OpenAlex)

Check 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
Use adaptive noise perturbation instead of classic SGLD (ablation)U-Net restoration trained with ASGLD · CS, Set11 and BSD68; tested noise and sampling ratiosPSNR (dB): 25.30 -> 26.02; 28.59 -> 29.35; 30.31 -> 31.11; 24.13 -> 24.56; 26.28 -> 27.16; 28.02 -> 28.75Improved
Use adaptive noise perturbation instead of classic SGLD (ablation)U-Net restoration trained with ASGLD · CS-MRI, Gaussian 1D/2D and Radial masksPSNR (dB): 25.11 -> 26.03; 26.25 -> 27.40; 26.30 -> 27.37Improved
Use adaptive noise perturbation instead of classic SGLD (ablation)U-Net phase retrieval · Phase retrieval, AWGN and Poisson noisePSNR (dB): 27.64 -> 28.42; 31.66 -> 32.42; 33.25 -> 34.82; 40.74 -> 42.04; 33.09 -> 35.85; 26.73 -> 29.15Improved
Use ASGLD rather than BNNSelf-supervised CS reconstruction · Set11 and BSD68, σ=10, 25% ratioPSNR (dB): 28.67 -> 29.35; 26.47 -> 27.16Improved
Use ASGLD rather than DIPSelf-supervised CS-MRI reconstruction · 1D Gaussian mask, no noisePSNR: 31.80 -> 32.18Improved
Use ASGLD rather than BNNSelf-supervised CS-MRI reconstruction · 2D Gaussian mask, σ=10%PSNR: 27.12 -> 27.40Improved
Use ASGLD rather than the strongest listed baselinePhase retrieval · Unnatural-6, AWGN SNR=15PSNR (dB): 32.81 -> 34.53Improved
Use ASGLD rather than the strongest listed baselinePhase retrieval · Natural-6, AWGN SNR=15PSNR (dB): 29.59 -> 30.30Improved
Use ASGLD rather than the strongest listed baselinePhase retrieval · Unnatural-6, Poisson α=9PSNR (dB): 39.41 -> 43.67Improved
Use ASGLD instead of SGLD (ablation)MRI reconstruction · MRI reconstruction with measurement noise, σ=25.5elapsed time: ≈640s -> ≈190sImproved
Use ASGLD instead of BNNMRI reconstruction · MRI reconstruction with measurement noise, σ=25.5elapsed time: ≈520s -> ≈190sImproved

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