AI Paper Atlas

Research map / 3D vision / Human pose estimation

Pose Partition Networks for Multi-Person Pose Estimation

Xuecheng Nie, Jiashi Feng, Junliang Xing, Shuicheng Yan

ECCV 2018 · 3D vision › Human pose estimation · approach: Multi-person pose estimation · 76 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
add joint partition (pose partition) before inference (ablation)PPN · MPII validationtotal AP: 78.6 -> 79.0Improved
add joint partition (pose partition) before inference (ablation)PPN · MPII validationinference time (ms): 3.4 -> 1.9Improved
add local greedy inference (ablation)PPN · MPII validationtotal AP: 77.8 -> 79.0Improved
add single-person pose refinement (ablation)PPN · MPII validationtotal AP: 76.4 -> 79.0Improved
test at 384x384 instead of 256x256 (ablation)PPN · MPII validationtotal AP: 76.3 -> 79.0Improved
add refinement and multi-scale testing to vanilla PPN (ablation)PPN · MPII validationtotal AP: 74.8 -> 79.0Improved
use 8 stacked hourglass stages instead of 1 (ablation)PPN · MPII validationtotal AP: 64.0 -> 79.0Improved
remove partition: head and shoulder detection vs per-joint (ablation)PPN · MPII validationhip and knee AP: Hip 77.8 -> 78.8, Knee 72.7 -> 73.1, Ankle 64.7 -> 65.7Hurt
replace Newell and Deng associative embedding with PPNbottom-up pose estimation · MPII multi-person (full test set)total AP: 77.5 -> 80.4Improved
replace Cao et al. with PPN for speedbottom-up pose estimation · MPII multi-person (full test set)runtime (s): 1.24 -> 0.77Improved
replace Xia et al. with PPNpose estimation · PASCAL-Person-Part (extended)total AP: 39.2 -> 46.6Improved
replace Insafutdinov et al. with PPNpose estimation · WAFtotal AP: 82.0 -> 84.8Improved

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