AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

Incomplete Gamma Kernels: Generalizing Locally Optimal Projection Operators

Patrick Stotko, Michael Weinmann, Reinhard Klein

TPAMI 2024 · Learning theory and algorithms › Kernels and regression theory · approach: RKHS kernel methods · 3 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
replace Gaussian localization kernel with actual K_LOP in WLOP density weights (simple scheme)WLOP · Bird irregular targetpoint cloud regularity sigma_Q: better than WLOP in 76.7% of combinationsImproved
solve full linear system for density weights instead of reciprocal densityWLOP simple scheme · Bird irregular targetpoint cloud regularity sigma_Q: outperforms WLOP in 99.3% and simple scheme in 98.7%Improved
use fitted kernel approximation parameters instead of CLOP parametersCLOP kernel approximation · K_LOP std ratio d=1std ratio: 0.7931 -> 0.9917Improved
use consistent approximation constraintkernel approximation · K_LOP std ratio d=4std ratio: 0.9640 -> 1.0440; unbiased at d to infinity (0.8881 -> 1)Mixed
use our kernel approximation instead of CLOP in WLOP smoothingWLOP smoothing · Block modelpoint-surface error at edgesImproved
apply optimal L1 scaling correction to kernel approximationkernel approximation · K_LOP L1 distanceL1 kernel errorImproved
replace L2 loss with localized LOP/Gaussian loss in normal filteringnormal filtering · Gargoyle and Boxfeature preservationImproved
replace Gaussian prior kernel with K_LOPTotalDenoising · ModelNet40 sigma=1.5%Chamfer distance (x1e-4): 0.466 -> 0.424Improved
replace Gaussian prior kernel with K_LOPTotalDenoising · ModelNet40 sigma=1%Chamfer distance (x1e-4): 0.343 -> 0.282Improved
replace Gaussian prior kernel with K_LOPTotalDenoising · ModelNet40 sigma=0.5%Chamfer distance (x1e-4): 0.173 -> 0.175No clear effect
replace Gaussian prior kernel with K_LOPTotalDenoising · Paris-rue-Madameoversmoothing and shrinkageImproved

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