AI Paper Atlas

Research map / Point clouds / Shape and surface completion

PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representations

Edgar Tretschk, Ayush Tewari, Vladislav Golyanik, Michael Zollhöfer, Carsten Stoll, Christian Theobalt

ECCV 2020 · Point clouds › Shape and surface completion · approach: Unsupervised shape correspondence · 87 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 patch-based representation instead of single global patchDeepSDF-style baseline · ShapeNet (100 shapes per category)IoU: 76.5 -> 92.1Improved
use patch-based representation instead of single global patchDeepSDF-style baseline · ShapeNet (100 shapes per category)F-score: 80.6 -> 94.8Improved
train on only Cabinet category: patches instead of global patchglobal-patch baseline · ShapeNet (trained on Cabinet)IoU: 66.4 -> 90.3Improved
cross-dataset test with patches instead of global patchglobal-patch baseline · Dynamic FAUST (trained on Cabinet)F-score: 25.0 -> 94.0Improved
increase number of patches from 3 to default (ablation)PatchNets · ShapeNet reduced test setIoU: 73.8 -> 91.6Improved
increase number of patches from 10 to default (ablation)PatchNets · ShapeNet reduced test setIoU: 85.2 -> 91.6Improved
increase patch latent/hidden size from 32 to default (ablation)PatchNets · ShapeNet reduced test setIoU: 82.8 -> 91.6Improved
increase latent/hidden size to 512 (ablation)PatchNets · ShapeNet reduced test setIoU: 91.6 -> 95.3Improved
train on full dataset instead of 100 shapes per category (ablation)PatchNets · ShapeNet reduced test setIoU: 91.6 -> 92.2Improved
add coverage loss L_cov (ablation)PatchNets · ShapeNet reduced test setIoU: 90.7 -> 91.6Improved
add rotation loss L_rot (ablation)PatchNets · ShapeNet reduced test setIoU: 92.5 -> 91.6Hurt
use normal-aligned rotation initialization with L_rot instead of random initialization (ablation)PatchNets · ShapeNet reduced test setIoU: 89.0 -> 91.6Improved

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