AI Paper Atlas

Research map / Point clouds / Shape and surface completion

3D Field of Junctions: A Noise-Robust, Training-Free Structural Prior for Volumetric Inverse Problems

Namhoon Kim, NARGES MOEINI, Justin Romberg, Sara Fridovich-Keil

ECCV 2026 · Point clouds › Shape and surface completion · approach: Point cloud diffusion generation · 0 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
increase patch size from 4 to 12 (stride 2) (ablation)3D FoJ · Engine CT P503D PSNR: 11.33 -> 17.70Improved
increase patch size from 4 to 12 (stride 2) (ablation)3D FoJ · Engine CT P50runtime: 7.2 min -> 24.7 minHurt
use stride 1 instead of stride 2 (patch size 8) (ablation)3D FoJ · Engine CT P50MS-SSIM: 0.796 -> 0.722Hurt
use stride 1 instead of stride 2 (patch size 8) (ablation)3D FoJ · Engine CT P50runtime: 13.1 min -> 47.3 minHurt
increase batch size from 1 to 6 (ablation)3D FoJ · Engine CT P50runtime: 37.6 min -> 17.2 minImproved
increase batch size from 1 to 6 (ablation)3D FoJ · Engine CT P503D PSNR: 17.17 -> 17.17No clear effect
use 3 regions per junction instead of 2 (ablation)3D FoJ · Engine CT P503D PSNR: 17.03 -> 17.17Mixed
increase regions per junction from 3 to 8 (ablation)3D FoJ · Engine CT P503D PSNR: 17.17 -> 16.12Hurt
use 3D FoJ prior instead of 3D-TVlow-dose CT reconstruction · Low-dose CT P50MS-SSIM: 0.703 -> 0.754Improved
use 3D FoJ instead of R2-Gaussianlow-dose CT reconstruction · Low-dose CT P10003D PSNR: 15.87 -> 16.72Hurt
use 3D FoJ instead of PointCleanNetpoint cloud denoising · Outlier noise 60%Chamfer-L2: 12.98 -> 0.22Improved
use 3D FoJ instead of PointCVaRpoint cloud denoising · Spread noise 100kChamfer-L2: 0.44 -> 0.17Improved

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