AI Paper Atlas

Research map / Point clouds / Shape and surface completion

GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions

Gene Chou, Ilya Chugunov, Felix Heide

NeurIPS 2022 · Point clouds › Shape and surface completion · approach: Distance-function surface reconstruction · 12 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 semi-supervised stage on top of meta-learning stage (ablation)GenSDF · Acronym unseen classesmean Chamfer Distance: 0.515 -> 0.407Improved
add meta-learning pretraining before semi-supervised stage (ablation)GenSDF · Acronym unseen classesmean Chamfer Distance: 0.889 -> 0.407Improved
add meta-learning stage on top of plain supervised training on 20 classes (ablation)supervised baseline · Acronym unseen classesmean Chamfer Distance: 1.164 -> 0.515Improved
episodic meta-learning vs supervised training on 76 classes (ablation)supervised baseline · Acronym unseen classesmean Chamfer Distance: 0.908 -> 0.515Improved
meta-learning vs supervised on seen classes (ablation)supervised baseline · Acronym seen classesmean Chamfer Distance: 0.695 -> 0.503Improved
add semi-supervised stage on seen classes (ablation)GenSDF · Acronym seen classesmean Chamfer Distance: 0.503 -> 0.351Improved
use signed nearest-neighbor self-supervised loss vs SAL/NeuralPullunsupervised SDF learning · Four classesmean Chamfer Distance: 2.419 -> 0.218Improved
add two-stage semi-supervised training over self-supervised only (ablation)GenSDF · Four classesmean Chamfer Distance: 0.218 -> 0.089Improved
increase Gaussian noise variance on input point cloudGenSDF · YCB PitcherChamfer Distance: 0.870 -> 5.208 (sigma^2 0 -> 0.1)Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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