AI Paper Atlas

Research map / 3D vision / Human pose estimation

ElePose: Unsupervised 3D Human Pose Estimation by Predicting Camera Elevation and Learning Normalizing Flows on 2D Poses

CVPR 2022 · 3D vision › Human pose estimation · approach: Monocular 2D-to-3D pose lifting · 40 citations (OpenAlex)

Check 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 the bone-length prior to the base loss (ablation)ElePose · Human3.6MPA-MPJPE: 77.9 -> 48.1Helped
Add elevation prediction to the base loss and bone-length prior (ablation)ElePose · Human3.6MPA-MPJPE: 48.1 -> 45.5Helped
Add PCA projection to the base loss and bone-length prior (ablation)ElePose · Human3.6MPA-MPJPE: 48.1 -> 43.1Helped
Add elevation prediction to the base loss, bone-length prior, and PCA projection (ablation)ElePose · Human3.6MPA-MPJPE: 43.1 -> 36.7Helped
Add PCA projection to the base loss, bone-length prior, and elevation prediction (ablation)ElePose · Human3.6MPA-MPJPE: 45.5 -> 36.7Helped
Use 22–30 PCA bases rather than a smaller or larger subspace (ablation)ElePose normalizing flow · Human3.6MPA-MPJPE: above 100mmMixed
Learn elevation angles and their dataset distribution rather than using a predefined elevation priorElePose · Human3.6MPA-MPJPE: approximately 15%Helped
Use learned elevation prediction and distributionElePose · Human3.6MMPJPE: more than 22%Helped

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