AI Paper Atlas

Research map / 3D vision / Face and shape

MDND: Unsupervised Learning Guided by Non-Differentiable Refinement for Shape Correspondence

Qinsong Li, Jing Meng, Haibo Wang, Shengjun Liu

AAAI 2026 · 3D vision › Face and shape · approach: Primitive-based shape abstraction · 1 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
use non-differentiable iterative refinement as supervisory oracle instead of a differentiable solver (ablation)MDND · TOPKIDSmean geodesic error x100: 9.4 -> 3.5Improved
use non-differentiable iterative refinement instead of differentiable solver (ablation)MDND · SMALmean geodesic error x100: 4.2 -> 3.1Improved
use non-differentiable iterative refinement instead of differentiable solver (ablation)MDND · SCAPEmean geodesic error x100: 2.0 -> 1.9Improved
use hybrid LBO+ELA basis in the refiner instead of LBO only (ablation)MDND oracle (HWF) · TOPKIDSmean geodesic error x100: 14.7 -> 3.5Improved
use hybrid LBO+ELA basis instead of ELA only (ablation)MDND oracle (HWF) · TOPKIDSmean geodesic error x100: 5.5 -> 3.5Improved
use hybrid basis instead of LBO only (ablation)MDND oracle (HWF) · SMALmean geodesic error x100: 4.9 -> 3.1Improved
use hybrid basis instead of LBO only (ablation)MDND oracle (HWF) · SCAPEmean geodesic error x100: 2.2 -> 1.9Improved
MDND versus state of the art on near-isometric FAUST and SCAPEMDND · FAUST and SCAPEmean geodesic error x100: FAUST 1.6, SCAPE 1.9 vs HybridFMaps 1.5, ULRSSM 1.8No clear effect
MDND on cross-dataset generalizationMDND · FAUST to SCAPEmean geodesic error x100: 6.4 -> 2.1 (vs ULRSSM)Improved
MDND on TOPKIDS compared with next best methodMDND · TOPKIDSmean geodesic error x100: 5.0 -> 3.5Improved

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