AI Paper Atlas

Research map / Image generation / Image restoration

Blind Image Deconvolution Using Variational Deep Image Prior

Dong Huo, Abbas Masoumzadeh, Rafsanjany Kushol, Yee‐Hong Yang

TPAMI 2023 · Image generation › Image restoration · approach: Image deblurring · 28 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
use variational Bayes with non-zero S(Is) instead of MAP, with extreme-channel prior (ablation)DIP-Extreme · Lai et al. syntheticPSNR: 22.39 -> 24.14Improved
use variational Bayes with non-zero S(Is) instead of MAP, with sparse prior (ablation)DIP-Sparse · Lai et al. syntheticPSNR: 20.53 -> 25.12Improved
use variational Bayes with non-zero S(Is) instead of MAP, with sparse prior (ablation)DIP-Sparse · Lai et al. syntheticSSIM: 0.659 -> 0.869Improved
add non-zero S(Is) without additive priors (ablation)DIP · Lai et al. syntheticPSNR: 20.67 -> 20.87No clear effect
add sparse image prior to DIP under MAP (ablation)DIP · Lai et al. syntheticPSNR: 20.67 -> 20.53Hurt
add extreme channel prior to DIP under MAP (ablation)DIP · Lai et al. syntheticPSNR: 20.67 -> 22.39Improved
add sparse prior with variational Bayes (VDIP-Sparse vs VDIP-Std) (ablation)VDIP · Lai et al. syntheticPSNR: 20.87 -> 25.12Improved
use VDIP-Sparse instead of VDIP-Extreme (sparse vs extreme-channel prior) (ablation)VDIP · Lai et al. syntheticPSNR: 24.14 -> 25.12Improved
use variational Bayes instead of MAP, sparse prior (ablation)DIP-Sparse · Lai et al. synthetickernel recovery error: 0.00144 -> 0.00092Improved
use variational Bayes instead of MAP, extreme prior (ablation)DIP-Extreme · Lai et al. synthetickernel recovery error: 0.00113 -> 0.00102Improved
use variational Bayes instead of MAP, sparse prior (ablation)DIP-Sparse · Lai et al. real blurredPIQE: 54.9295 -> 34.3614Improved
use variational Bayes instead of MAP, sparse prior (ablation)DIP-Sparse · Lai et al. real blurredNIQE: 7.9063 -> 3.8882Improved

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