AI Paper Atlas

Research map / 3D vision / Human pose estimation

G2L-Net: Global to Local Network for Real-Time 6D Pose Estimation With Embedding Vector Features

Wei Chen, Xi Jia, Hyung Jin Chang, Jinming Duan, Ales Leonardis

CVPR 2020 · 3D vision › Human pose estimation · approach: Category-level pose estimation · 119 citations (OpenAlex)

Check 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
Add 3D sphere localization (ablation)G2L-Net · LINEMODADD(-S) accuracy: 93.4% -> 95.8%Improved
Add point-wise embedding vector features (EVF) (ablation)G2L-Net · LINEMODADD(-S) accuracy: 95.8% -> 98.4%Improved
Add rotation residual estimator (RRE) (ablation)G2L-Net · LINEMODADD(-S) accuracy: 98.4% -> 98.7%Improved
Vary keypoint selection among 8 bounding-box corners and FPS sets (ablation)G2L-Net · LINEMODADD(-S) accuracy: 98.7% -> 98.5%; 98.4%; 98.6%No clear effect
Use G2L-Net instead of the strongest listed baselineLINEMOD · LINEMODADD(-S) accuracy: 96.3 % -> 98.7 %Improved
Use G2L-Net instead of MCNYCB-Video · YCB-VideoADD-S AUC: 94.3 % -> 92.4 %Hurt

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