AI Paper Atlas

Research map / Image generation / Image restoration

Fractal Dimension Invariant Filtering and Its CNN-Based Implementation

Hongteng Xu, Junchi Yan, Nils Persson, Weiyao Lin, Hongyuan Zha

CVPR 2017 · Image generation › Image restoration · approach: Inverse imaging with learned priors · 23 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
apply unsupervised FDIF filtering with thresholding instead of traditional non-learning detectorscurve detection · AFM image benchmarkODS: 0.058 -> 0.717; 0.629 -> 0.717Improved
use supervised FraCNN features with logistic regression rather than HEDcurve detection · AFM image benchmarkODS: 0.722 -> 0.743Improved
use supervised FraCNN features with logistic regression rather than HEDcurve detection · AFM image benchmarkAP: 0.784 -> 0.730Hurt
increase FDIF iterations from 3 to 6 (ablation)FDIF + BP · AFM image benchmarkODS: 0.717 -> 0.715Hurt
increase FDIF iterations from 3 to 6 (ablation)FDIF + LR · AFM image benchmarkOIS: 0.770 -> 0.767Hurt
increase FraCNN depth from 6 to 12 (ablation)FraCNN + BP · AFM image benchmarkODS: 0.691 -> 0.689Hurt
increase FraCNN depth from 6 to 12 (ablation)FraCNN + LR · AFM image benchmarkAP: 0.730 -> 0.718Hurt
replace adaptive direction selection with a predefined anisotropic filter bankFDIF curve detector · AFM image benchmarkAP: 0.699 -> 0.708Mixed
use FDIF rather than HED on Brodatz texture curvescurve detection · Brodatz texture imagesODS: 0.518 -> 0.530Improved
use FDIF rather than HED on Brodatz texture curvescurve detection · Brodatz texture imagesAP: 0.791 -> 0.803Improved

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