AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Iterative Multiplicative Filters for Data Labeling

Ronny Bergmann, Jan Henrik Fitschen, Johannes Persch, Gabriele Steidl

IJCV 2017 · Learning theory and algorithms › Matrix and tensor methods · approach: Randomized sketching for matrices · 16 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
enlarge the uniform local neighborhood size (ablation)iterative multiplicative filter · Mandrill labelingiterations to stop: 27 -> 39Hurt
use nonlocal instead of local weights (ablation)iterative multiplicative filter · Mandrill labelingdetail preservation (qualitative)Improved
set epsilon to 0 instead of a small positive value (ablation)iterative multiplicative filter · 1D signal labeling examplelabel assignment: eps 10^-10 gives labels 1 1 3 3 3 3 3 2 2 2; eps 0 gives all labels 3Hurt
add extra factor A_i from the earlier scheme (ablation)iterative multiplicative filter · Mandrill labelinglabeled pixels: differ only in 4 percent of the image pixelsNo clear effect
use multiplicative labeling instead of Gaussian-smoothing simple labelingmandrill labeling · Mandrill labelingdetail sharpness (qualitative)Improved
use multiplicative labeling instead of variational methodmandrill labeling · Mandrill labelingdetail (qualitative)No clear effect
use nonlocal instead of equal weights for DT-MRI labeling (ablation)iterative multiplicative filter · DT-MRI Camino slice z=35boundary and diffusion path capture (qualitative)Improved
use the labeling approach instead of thresholding grain boundariesEBSD magnesium data · EBSD Magnesiumgrain boundary smoothness (qualitative)Improved

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