AI Paper Atlas

Research map / Image generation / Super-resolution

Deblurring Natural Image Using Super-Gaussian Fields

Yuhang Liu, Wenyong Dong, Dong Gong, Lei Zhang, Qinfeng Shi

ECCV 2018 · Image generation › Super-resolution · approach: Efficient lightweight super-resolution · 30 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 super-Gaussian fields instead of GSM-FoE (Schmidt et al.) as the high-order MRF priornon-blind deblurring · Levin et al. dataset (non-blind)average SSD: 25.43 -> 21.77Improved
use super-Gaussian fields instead of GSM-FoE (Schmidt et al.) as the high-order MRF priornon-blind deblurring · Levin et al. dataset (non-blind)run time (s): >10000 -> 485Improved
use super-Gaussian fields instead of Zoran and Weiss priornon-blind deblurring · Levin et al. dataset (non-blind)average SSD: 24.60 -> 21.77Improved
use super-Gaussian fields instead of Zoran and Weiss priornon-blind deblurring · Levin et al. dataset (non-blind)run time (s): 3093 -> 485Improved
use super-Gaussian fields instead of the dark channel prior (Pan et al.)blind deblurring · Levin and Kohler datasets (blind)success percent / PSNRNo clear effect
use super-Gaussian fields instead of the dark channel prior (Pan et al.)blind deblurring · Challenging example from Kohler et al. (blind)PSNR: 26.259 -> 27.5427Improved
add filter updating (learn filters from the blurred image at each scale) (ablation)super-Gaussian fields deblurring · Levin et al. dataset (blind)success percent vs error ratioImproved
use adaptive learned filters instead of the gradient filters (ablation)super-Gaussian fields deblurring · qualitative example (Fig. 3)sparsity of response / visual sharpnessImproved
use super-Gaussian fields instead of the pre-learned GSM-FoE prior in blind deblurring (ablation)coarse-to-fine blind deblurring · Levin et al. dataset (blind)success percent vs error ratioImproved

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