AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Weakly Supervised Deep Functional Maps for Shape Matching

Abhishek Sharma, Maks Ovsjanikov

NeurIPS 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory

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
remove weak rigid alignment of training data (ablation)Ours (All losses) · Surreal->Scapegeodesic error x100: 8.3 -> 22Hurt
remove weak rigid alignment of training data (ablation)Ours (All losses) · Surreal->Faustgeodesic error x100: 5.0 -> 8.0Hurt
use only E3 Laplacian commutativity loss instead of all losses (ablation)Ours · Surreal->Scapegeodesic error x100: 8.3 -> 10.5Hurt
use only E1 loss instead of all losses (ablation)Ours · Surreal->Scapegeodesic error x100: 8.3 -> 13Hurt
use only E2 orthonormality loss instead of all losses (ablation)Ours · Surreal->Scapegeodesic error x100: 8.3 -> 16Hurt
use E1+E3 instead of all three losses (ablation)Ours · Surreal->Faustgeodesic error x100: 5.0 -> 6.3Hurt
add ZoomOut refinementOurs · Faust Fgeodesic error x100: 3.3 -> 1.9Improved
add ZoomOut refinementOurs · Surreal->Scapegeodesic error x100: 8.3 -> 5.5Improved
weakly supervised method vs fully supervised GeomFmap+zoOurs + zo · Surreal->Scapegeodesic error x100: 9.2 -> 5.5Improved
weakly supervised method vs fully supervised GeomFmap+zoOurs + zo · Surreal->Faustgeodesic error x100: 2.5 -> 2.8Hurt
train on Scape and test on Faust vs GeomFmap+zoOurs + zo · S on Fgeodesic error x100: 4.3 -> 4.3No clear effect
use GeomFmap supervised loss with our frameworkGeomFmap +Zo · Surreal->Scapegeodesic error x100: 9.2 -> 4.6Improved

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