AI Paper Atlas

Research map / Image generation / Image restoration

Learning Spatially-Variant MAP Models for Non-Blind Image Deblurring

Jiangxin Dong, Stefan Roth, Bernt Schiele

CVPR 2021 · Image generation › Image restoration · approach: Image deblurring · 32 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
jointly learn image-adaptive data and regularization terms (ablation)SVMAP · Martin et al. [24], 1% Gaussian noisePSNR (dB): 29.88 -> 31.89; 31.59 -> 31.89Improved
jointly learn image-adaptive data and regularization terms (ablation)SVMAP · Saturated-pixel datasetPSNR (dB): 30.58 -> 33.91; 29.47 -> 33.91Improved
add spatially-variant filters instead of spatially-invariant filters (ablation)SVMAP · Martin et al. [24], 1% Gaussian noisePSNR (dB)/SSIM: 31.25/0.8861 -> 31.89/0.8973Improved
add spatially-variant filters instead of spatially-invariant filters (ablation)SVMAP · Martin et al. [24], 5% Gaussian noisePSNR (dB)/SSIM: 26.72/0.7302 -> 27.25/0.7550Improved
jointly learn data and regularization terms with spatially-variant featuresSVMAP · Martin et al. [24], 1% noisePSNR (dB): 30.92 -> 31.89Improved
jointly learn data and regularization terms with spatially-variant featuresSVMAP · Martin et al. [24], 5% noisePSNR (dB): 25.93 -> 27.25Improved
jointly learn data and regularization terms with spatially-variant featuresSVMAP · Sun et al. [41], 1% noisePSNR (dB): 33.57 -> 34.51Improved
jointly learn data and regularization terms with spatially-variant featuresSVMAP · Sun et al. [41], 5% noisePSNR (dB): 27.75 -> 29.20Improved
jointly learn data and regularization terms with spatially-variant featuresSVMAP · Saturated-pixel datasetPSNR (dB): 33.02 -> 33.91Improved

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