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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| enlarge the uniform local neighborhood size (ablation) | iterative multiplicative filter · Mandrill labeling | iterations to stop: 27 -> 39 | Hurt |
| use nonlocal instead of local weights (ablation) | iterative multiplicative filter · Mandrill labeling | detail preservation (qualitative) | Improved |
| set epsilon to 0 instead of a small positive value (ablation) | iterative multiplicative filter · 1D signal labeling example | label assignment: eps 10^-10 gives labels 1 1 3 3 3 3 3 2 2 2; eps 0 gives all labels 3 | Hurt |
| add extra factor A_i from the earlier scheme (ablation) | iterative multiplicative filter · Mandrill labeling | labeled pixels: differ only in 4 percent of the image pixels | No clear effect |
| use multiplicative labeling instead of Gaussian-smoothing simple labeling | mandrill labeling · Mandrill labeling | detail sharpness (qualitative) | Improved |
| use multiplicative labeling instead of variational method | mandrill labeling · Mandrill labeling | detail (qualitative) | No clear effect |
| use nonlocal instead of equal weights for DT-MRI labeling (ablation) | iterative multiplicative filter · DT-MRI Camino slice z=35 | boundary and diffusion path capture (qualitative) | Improved |
| use the labeling approach instead of thresholding grain boundaries | EBSD magnesium data · EBSD Magnesium | grain boundary smoothness (qualitative) | Improved |
Closest papers by meaning
- Minimizing the Maximal Rank (CVPR 2016 · similarity 0.84)
- Scene Labeling Using Sparse Precision Matrix (CVPR 2016 · similarity 0.84)
- Sublabel-Accurate Convex Relaxation of Vectorial Multilabel Energies (ECCV 2016 · similarity 0.84)
- Continuous Partitioning for Graph-Based Semi-Supervised Learning (NeurIPS 2024 · similarity 0.83)
- High Rank Matrix Completion With Side Information (AAAI 2018 · similarity 0.83)
- Dimensionality Reduction on SPD Manifolds: The Emergence of Geometry-Aware Methods (TPAMI 2017 · similarity 0.83)
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.
