AI Paper Atlas

Research map / 3D vision / Human pose estimation

Adaptive Hypergraph Neural Network for Multi-Person Pose Estimation

AAAI 2022 · 3D vision › Human pose estimation · approach: Multi-person pose estimation · 12 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 AP-HNN to HRNet (ablation)HRNet · COCO test-dev2017AP: 74.9 -> 76.3Helped
Add AP-HNN to SBN (ablation)SBN · COCO test-dev2017AP: 73.7 -> 75.1Helped
Add SIC to the initial pose estimator (ablation)AD-HNN initial pose estimator · COCO val2017AP: 74.3 -> 74.8Helped
Add AP-HNN to the initial pose estimator (ablation)AD-HNN initial pose estimator · COCO val2017AP: 74.3 -> 75.9Helped
Use SIC and AP-HNN together (ablation)AD-HNN · COCO val2017AP: 74.3 -> 76.4Helped
Replace the hypergraph with a fixed tree-like GNN (ablation)AD-HNN · COCO val2017AP: 75.3 -> 74.9Hurt
Keep joint relations fixed rather than adapt them (ablation)AD-HNN · COCO val2017AP: 75.9 -> 75.3Hurt
Use dynamic k-NN connections instead of learnable adaptive hypergraph relations (ablation)AD-HNN · COCO val2017AP: 75.9 -> 75.6Hurt
Add AP-HNN to the pose estimator (ablation)AD-HNN · COCO val2017 self-occlusion subsetAP: 76.0 -> 79.5Helped

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