atlas

Research map / 3D vision / Human pose estimation

LCR-Net: Localization-Classification-Regression for Human Pose

CVPR 2017 · 3D vision › Human pose estimation · approach: Monocular 2D-to-3D pose lifting · 317 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
Integrate neighboring pose proposals with score-weighted averaging instead of selecting proposals with NMS (ablation)LCR-Net · Human3.6M P23D pose error (mm), absolute: 89.8 -> 87.7Helped
Integrate neighboring pose proposals instead of selecting proposals with NMS (ablation)LCR-Net · Human3.6M P13D pose error (mm), absolute: 65.9 -> 63.2Helped
Integrate neighboring pose proposals instead of selecting proposals with NMS (ablation)LCR-Net · Human3.6M P33D pose error (mm), aligned: 73.1 -> 72.7Helped
Use 100 clustered anchor-poses rather than a single pose class (ablation)LCR-Net · Human3.6M P13D pose error (mm), NMS: 74.2 -> 70.7Helped
Increase the number of anchor-poses from 100 to 500 (ablation)LCR-Net · Human3.6M P13D pose error (mm), NMS: 70.7 -> 85.6Hurt
Increase the number of anchor-poses from 100 to 200 (ablation)LCR-Net · Human3.6M P13D pose error (mm), NMS: 70.7 -> 77.0Hurt
Add annotated COCO images to the training set (ablation)LCR-Net · MPII validationPCKh@0.5: 78.5%Helped
Use PPI rather than NMS for pose proposal postprocessing (ablation)LCR-Net · MPII validationPCKhHelped

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