AI Paper Atlas

Research map / 3D vision / Face and shape

3D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Data

Maurice Weiler, Mario Geiger, Max Welling, Wouter Boomsma, Taco S Cohen

NeurIPS 2018 · 3D vision › Face and shape · approach: Learned mesh and geometric representations · 248 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
replace conventional convolutions with SE(3)-equivariant convolutionsAmino acid environment prediction · Amino acid environments test setaccuracy: 0.50 -> 0.58Improved
use SE(3)-equivariant convolutions rather than a concentric-grid convolutionAmino acid environment prediction · Amino acid environments test setaccuracy: 0.56 -> 0.58Improved
use an SE(3)-equivariant CNN rather than a conventional CNNTetris classification · Tetris rotated test setaccuracy: 27±7% -> 99±2%Improved
enable low-pass filtering before downsampling (ablation)3D Steerable CNN · Tetris rotated test setaccuracy: 36% ± 6% -> 99% ± 2%Improved
enable low-pass filtering before downsampling (ablation)Conventional CNN · Tetris rotated test setaccuracy: 24% ± 4% -> 27% ± 7%No clear effect
use gated nonlinearities instead of norm nonlinearities (ablation)3D Steerable CNN · Experiments reported in the paperconvergence speedImproved
replace stride-based downsampling with smoothing before subsampling (ablation)3D Steerable CNN · Spatial dimension reductionrotational equivarianceImproved
use the 3D Steerable CNN rather than a conventional CNNCATH protein architecture classification · CATH test setaccuracyImproved
use the proposed 3D Steerable CNN rather than the strongest listed SHREC17 baselineSHREC17 3D model classification · SHREC17total score: 1.13 -> 1.11Hurt

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