AI Paper Atlas

Research map / Image generation / Image restoration

Representing Noisy Image Without Denoising

Shuren Qi, Yushu Zhang, Chao Wang, Tao Xiang, Xiaochun Cao, Yong Xiang

TPAMI 2024 · Image generation › Image restoration · approach: Raw and low-light noise modeling · 10 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
move moment computation from image domain to Radon domain (FM to FMR)fractional-order moments · COREL, K=20, noise variance 0.2correct classification (%): 80.14 -> 95.72Improved
move moment computation from image domain to Radon domain (average over parameter settings) (ablation)fractional-order moments · COREL, noise variance 0.2correct classification (%): 72.34 -> 85.02Improved
move moment computation to Radon domain, best resultsfractional-order moments · COREL, noise variance 0.3correct classification (%): 45.31 -> 58.5Improved
move FJFM(3,3,2) to Radon domainfractional-order moments · COREL, K=10, noise variance 0.3correct classification (%): 36.75 -> 33Hurt
use Radon projection for reconstruction instead of image domain (ablation)harmonic FM · reconstruction of noisy Fourier imageMSRE: 0.060 -> 0.046Improved
add noise augmentation to rotation augmentationResNet-50 · COREL, noise variance 0.1correct classification (%): 1.28 -> 66.33Improved
add noise augmentation to rotation augmentationGoogLeNet · COREL, noise variance 0.1correct classification (%): 2.22 -> 52.47Improved
add noise augmentation to rotation augmentationPCANet · COREL, noise variance 0.1correct classification (%): 23.22 -> 13.72Hurt
add denoising preprocessing to rotation augmentationResNet-50 · COREL, noise variance 0.1correct classification (%): 1.28 -> 23.78Improved
add denoising preprocessing to rotation augmentationResNet-50 · COREL, noise variance 0.3correct classification (%): 1.25 -> 4Improved
add rotation augmentation to naive trainingPCANet · COREL, clean imagescorrect classification (%): 19.64 -> 100Improved
use FMR instead of rotation-invariant SIDtemplate matching · fluorescence microscopy, rotation and noiseaccuracy (%): 20 -> 93Improved

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