AI Paper Atlas

Research map / 3D vision / Human pose estimation

Look into Person: Joint Body Parsing & Pose Estimation Network and a New Benchmark

Xiaodan Liang, Ke Gong, Xiaohui Shen, Liang Lin

TPAMI 2018 · 3D vision › Human pose estimation · approach: Heatmap and coordinate pose encoding · 368 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 multi-scale features (ablation)JPPNet · LIP test parsingmean IoU: 49.48 -> 50.83Improved
add one-stage refinement (ablation)JPPNet · LIP test parsingmean IoU: 49.48 -> 49.58No clear effect
add one-stage refinement on top of multi-scale features (ablation)JPPNet · LIP test parsingmean IoU: 50.83 -> 51.36Improved
add second refinement stage (ablation)JPPNet · LIP test parsingmean IoU: 51.36 -> 50.64Hurt
add multi-scale features (ablation)JPPNet · LIP test pose (JPPNet ablation)total PCKh: 79.6 -> 80.9Improved
add one-stage refinement (ablation)JPPNet · LIP test pose (JPPNet ablation)total PCKh: 79.6 -> 82.1Improved
add multi-scale features on top of one-stage refinement (ablation)JPPNet · LIP test pose (JPPNet ablation)total PCKh: 82.1 -> 82.5Improved
add second refinement stage (ablation)JPPNet · LIP test pose (JPPNet ablation)total PCKh: 82.5 -> 82.7Improved
add self-supervised structure-sensitive loss to AttentionSS-JPPNet · LIP val parsingmean IoU: 42.92 -> 44.73Improved
add self-supervised structure-sensitive loss to AttentionSS-JPPNet · PASCAL-Person-Partmean IoU: 56.39 -> 59.36Improved
add joint (pose) context to parsing instead of DeepLab ResNet-101 aloneJPPNet · LIP val parsingmean IoU: 44.80 -> 51.37Improved
add parsing context to pose estimation vs HourglassJPPNet · LIP test posetotal PCKh: 77.6 -> 82.7Improved

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