AI Paper Atlas

Research map / 3D vision / Human pose estimation

3D Human Pose Estimation Using Möbius Graph Convolutional Networks

Niloofar Azizi, Horst Possegger, Emanuele Rodolà, Horst Bischof

ECCV 2022 · 3D vision › Human pose estimation · approach: Transformer-based multi-view 3D pose · 27 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 Mobius-transformed complex spectral filters instead of real Chebychev spectral GCNMobiusGCN · Human3.6M Protocol #1 (2D GT input)MPJPE (mm): 110.6 -> 40.0Improved
MobiusGCN 0.04M vs SemGCN 0.43M with 2D GT inputMobiusGCN · Human3.6M Protocol #1 (2D GT input)MPJPE (mm): 43.8 -> 40.0Improved
MobiusGCN 0.04M vs SemGCN 0.43M with 2D HG inputMobiusGCN · Human3.6M Protocol #1 (2D HG input)MPJPE (mm): 60.8 -> 54.2Improved
MobiusGCN 0.16M vs light GraphSH 0.44MMobiusGCN · Human3.6M Protocol #1 (2D GT input)MPJPE (mm): 39.2 -> 36.2Improved
increase channels from 64 to 128 (0.04M to 0.16M parameters), 2D GT inputMobiusGCN · Human3.6M Protocol #1 (2D GT input)MPJPE (mm): 40.0 -> 36.2Improved
increase channels from 64 to 128 (0.04M to 0.16M parameters), 2D HG inputMobiusGCN · Human3.6M Protocol #1 (2D HG input)MPJPE (mm): 54.2 -> 52.1Improved
MobiusGCN 0.16M vs GraphSH 3.7M, 2D HG inputMobiusGCN · Human3.6M Protocol #1 (2D HG input)MPJPE (mm): 51.9 -> 52.1No clear effect
increase training subjects from S1 to S1+S5 (ablation)MobiusGCN · Human3.6M training subjects (2D GT input)MPJPE (mm): 62.3 -> 47.9Improved
increase training subjects from S1+S5 to S1+S5+S6 (ablation)MobiusGCN · Human3.6M training subjects (2D GT input)MPJPE (mm): 47.9 -> 43.1Improved
MobiusGCN vs GraphSH on MPI-INF-3DHP outdoor sceneMobiusGCN · MPI-INF-3DHP Outdoor3D PCK: 75.2 -> 83.1Improved
MobiusGCN vs GraphSH on MPI-INF-3DHP GS and noGS scenesMobiusGCN · MPI-INF-3DHP GS / noGS3D PCK: 81.5 -> 79.2 (GS); 81.7 -> 77.3 (noGS)Hurt
MobiusGCN trained on one subject vs semi-supervised methods using unlabeled data, 2D HG inputMobiusGCN · Human3.6M semi-supervised comparison (2D HG input)MPJPE (mm): 88.8 -> 82.3Improved

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