AI Paper Atlas

Research map / Image generation / Image restoration

Learn From Unpaired Data for Image Restoration: A Variational Bayes Approach

Dihan Zheng, Xiaowen Zhang, Kaisheng Ma, Chenglong Bao

TPAMI 2022 · Image generation › Image restoration · approach: Efficient restoration architectures · 31 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 clean-to-noisy (C2N) instead of noisy-to-clean (N2C) data generation (ablation)LUD-VAE · SIDD benchmarkPSNR: 34.74 -> 34.82Improved
add clean-to-noisy (C2N) instead of noisy-to-clean (N2C) data generation (ablation)LUD-VAE · SIDD benchmarkSSIM: 0.901 -> 0.926Improved
add pre-processing Gaussian noise (sigma_x=15, sigma_y=10) versus none (ablation)LUD-VAE · AIM19PSNR: 21.66 -> 22.32Improved
add pre-processing Gaussian noise (sigma_x=15, sigma_y=10) versus none (ablation)LUD-VAE · AIM19MMD: 7.64 -> 2.34Improved
increase stochastic layers from 1 to 3 (ablation)LUD-VAE · AIM19PSNR: 21.24 -> 22.32Improved
increase stochastic layers from 3 to 5 (ablation)LUD-VAE · AIM19PSNR: 22.32 -> 22.07No clear effect
replace gamma-correction-only degradation with LUD-VAE learned degradation (ablation)LLFlow · LOLPSNR: 22.87 -> 24.92Improved
replace bicubic with nearest downsampling (ablation)LUD-VAE · AIM19PSNR: 22.32 -> 22.65Improved
replace bicubic with bilinear downsampling (ablation)LUD-VAE · AIM19PSNR: 22.32 -> 21.62Hurt
add LUD-VAE degradation instead of bicubic-only training dataESRGAN · AIM19PSNR: 21.69 -> 22.32Improved
use LUD-VAE instead of DeFlow-NP (no pretrained network)ESRGAN · NTIRE20SSIM: 0.6777 -> 0.7178Improved
use LUD-VAE instead of AWGN synthetic noiseDnCNN · SIDD benchmarkPSNR: 32.12 -> 34.82Improved

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