AI Paper Atlas

Research map / 3D vision / Face and shape

TetraSDF: Analytic Isosurface Extraction with Multi-resolution Tetrahedral Grid

Seonghun Oh, Youngjung Uh, Jin-Hwa Kim

ECCV 2026 · 3D vision › Face and shape · approach: Neural implicit surface reconstruction · 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
add input preconditioner A* (ablation)TetraSDF · Thingi10K (Large)Chamfer DistanceImproved
add encoder-derived preconditioner A* (ablation)HashGrid · Thingi10K (Large)Chamfer DistanceHurt
replace ReLU MLP (AM) with tetrahedral encoder + small ReLU MLPanalytic extraction · Stanford (Large)Chamfer Distance x1e-6: 5475 -> 1718Improved
replace trilinear HashGrid encoder (TropicalNeRF) with tetrahedral barycentric encoderanalytic extraction · Stanford (Small)Chamfer Distance x1e-6: 5367 -> 4210Improved
replace trilinear HashGrid encoder (TropicalNeRF) with tetrahedral barycentric encoderanalytic extraction · Stanford (Large)Chamfer Distance x1e-6: 1737 -> 1718Improved
replace TropicalNeRF extraction with exact analytic extractionTetraSDF · Thingi10KSSDF x1e-6: 144.1 -> 0.078Improved
analytic extraction instead of Marching Cubes at 1024TetraSDF network · Thingi10KSSDF x1e-6: 17.6 -> 0.078Improved
increase Marching Cubes resolution from 256 to 1024TetraSDF network · Thingi10KSSDF x1e-6: 217.0 -> 17.6Improved
use tetrahedral encoder instead of HashGridSDF fitting · Stanford (Large)GT-mesh SDF MAE: 0.53 -> 0.62No clear effect
use tetrahedral encoder instead of HashGridSDF fitting · Stanford (Small)GT-mesh SDF MAE: 3.24 -> 2.82Improved
enlarge plain ReLU MLP for Analytic Marching from H10/W90 to H20/W200AM · StanfordChamfer Distance x1e-6: 5475 -> 5062No clear effect
use TetraSDF instead of adaptive octree extractorsextraction on Large network · StanfordSSDF x1e-6: 8.85 -> 0.076Improved

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